Embed Size (px)
Citation preview
An Ultra-Wideband Location
Positioning Navigation System
for Mobile Robotics
James Knox (B0036062)
Supervisor : Dr J Condell
Final Report (EEE521M4)
April 2008
Abstract
There is considerable interest in the development of real time location systems
(RTLS) for industry. The ability to track the real-time location and movement of items
or people offers a range of useful applications in areas such as safety, security and
the supply chain. Current location determination technologies have limitations that
heavily restrict how and where these applications are implemented, including the
cost, accuracy of the location, calculation and inherent properties of the system.
Real-time Ultra Wide Band Location determination is a technology that uses Ultra-
wideband (UWB) to locate with precision a device in space and time. This real time
location system can deliver very high positional accuracy in harsh industrial
environments that previously caused problems for traditional location systems based
on Wi-Fi of RFID due to interference. Ultra Wide Band systems work by being able to
calculate the location of tags which are designed to be mounted on assets or to be
worn by a person. They transmit UWB signals that are received by sensors which
contain an array of antenna and ultra-wideband radio receivers. The data from these
sensors combined with dedicated software uses algorithms to work out the angle of
arrival (AOA) of the UWB signal from the tag which is then compared to the time
difference of arrival (TDOA). This information is determined between pairs of sensors
connected by timing cables. The combination of AOA and TDOA measurement
technologies delivers a precise three dimensional location system that is powerful,
reliable, robust - specifically designed for harsh industrial environments.
This new technology can locate with precision items in 3D space and opens up a
new world in interactive applications especially within the robotics industry. This
thesis examines the use of ultra wide band technology in tracking robots.
2
Table of Contents
Abstract...............................................................................................................................2Table of Contents...............................................................................................................3Table of Figures.................................................................................................................5Tables..................................................................................................................................7Graphs.................................................................................................................................8Acknowledgments............................................................................................................101 Introduction...............................................................................................................11
1.1 Aims......................................................................................................................111.2 Objectives.............................................................................................................121.3 Overview of Thesis............................................................................................12
2 Literature Review.....................................................................................................142.1 Global Positioning Systems..........................................................................142.2 Indoor Positioning...........................................................................................16
2.2.1 Wi-Fi (802.11) Based Indoor Positioning.................................................162.2.3 Ultra-Wideband location............................................................................18
2.3 Robotics and Location Determination..........................................................202.3.1 Robotic Navigation Sensors.......................................................................21
2.7 Conclusion.........................................................................................................233 Requirements Specification and Analysis..............................................................24
3.1 Problem Specification.....................................................................................243.2 System Scenario..............................................................................................243.3 Requirements Analysis...................................................................................24
3.3.1 Functional Requirements........................................................................24Client Side.................................................................................................................24Server Side...............................................................................................................253.3.2 Non-Functional Requirements...............................................................25
3.4 Hardware and Software Requirements..........................................................253.5 Development Languages..................................................................................253.6 Conclusion.........................................................................................................26
4 System Functionality: Order of Proceedings.......................................................274.1 HCI principles.....................................................................................................294.2 Interface Design.................................................................................................294.3 Platform Control and Location Engine Configuration...................................304.4 Sensor properties...............................................................................................304.5 Conclusion...........................................................................................................31
5 Design and Implementation...................................................................................325.1 System Structure...............................................................................................325.2 Implementation...................................................................................................335.3 DHCP Software..................................................................................................34
5.3.1 DHCP Log.....................................................................................................355.4 Location Engine Configuration.........................................................................365.5 Calibration of Sensors.......................................................................................375.6 Sensors scanning area plan.............................................................................385.7 Transmitting tags...............................................................................................395.8 Location sensors in operation..........................................................................40
3
5.9 Conclusion..........................................................................................................436 Testing.......................................................................................................................44
6.1 Location Testing.................................................................................................446.2 Calibration Testing.............................................................................................446.3 Error Tracking Testing.......................................................................................456.5 Conclusion...........................................................................................................47
7 Evaluation..................................................................................................................487.2 Evaluation Results.............................................................................................507.3 Evaluation results of tag location test.............................................................507.4 Evaluation results of water test........................................................................517.5 Evaluation results of material test...................................................................527.6 Evaluation results of other tests carried out..................................................537.3 Conclusion...........................................................................................................57
8 Project conclusion........................................................................................................589 Potential Future Work..................................................................................................60References........................................................................................................................61Appendix A........................................................................................................................64Test results on how a liquid can effect a tags location and detection.....................64Appendix B........................................................................................................................69Test results on how different materials effect a tags location and detection.........69Appendix C.......................................................................................................................81Results from tracking a person wearing a tag from point A to point B....................81Appendix D.......................................................................................................................86Results of tests carried out on the accuracy readings of a tags location................86Appendix E........................................................................................................................93Test results: Liquid disturbance test.............................................................................93Appendix F........................................................................................................................95Test Results: Three tags transmitting at the one time...............................................95Appendix G.......................................................................................................................97Configuration and setup of the location detection system.........................................97Appendix H.......................................................................................................................98Problems with the setup and configuration of the location detection system........98Appendix I.......................................................................................................................100
4
Table of FiguresFigure 1: GPS Satellites.................................................................................................15Figure 2: Trilateration. Wikipedia (2008)......................................................................15Figure 3: Vicon animation...............................................................................................17Figure 4: Vicon motion capture......................................................................................17Figure 5: Ubisense dealing with Multipath effects (Ubisense, 2008).......................18Figure 6: Ubisense compact tags, single sensors and series 7000 sensors.........18Figure 7: Car production line can be zoned using Ubisense (2008).......................19Figure 8: Bluetooth WI-FI and UWB Protocols (Shahril, 2008)................................20Figure 9: Wireless acoustic location with room-level resolution using ultrasound 21Figure 10: Beacon navigation (Lee, 2008)...................................................................22Figure 11: Beacon aided navigation for Robots..........................................................22Figure 12: .NET Framework (Woods, 2008)...............................................................26Figure 13: Use case diagram showing the functionality of the System...................27Figure 14: State diagram for a Location Detection System .......................28Figure 15: High-Level Sequence Diagram for Tracking RF tags.............................28Figure 16: Good Design Principles...............................................................................29Figure 17: Platform Control............................................................................................30Figure 18: Deployment of sensors................................................................................30Figure 19: Properties of the master sensor location..................................................31Figure 20: Evolutionary Development..........................................................................32Figure 21: Sensors wiring connection system.............................................................33Figure 22: Diagram showing the Location Detection System Setup.......................33Figure 23: IP Configuration settings of Laptop............................................................34Figure 24: IP Pool address edited to range from 192.168.3.7-10............................34Figure 25: The DHCP Server software.........................................................................35Figure 26: Ping request successful, confirms a working network connection........35Figure 27: DHCP log in Notepad...................................................................................36Figure 28: List of Dynamic addresses and the Physical addresses of sensors.....36Figure 29: Location Engine Configuration – Sensor Status......................................37Figure 30: Location Engine Configuration – Sensor and Cells.................................37Figure 31: Calibration of each sensor in progress......................................................38Figure 32: Thresholds set for the four sensors...........................................................38Figure 33: Sensors scanning area to be covered.......................................................39Figure 34: Internal components of a Transmitting tag...............................................40Figure 35: Location of Tag at ground level at the centre of area plan....................40Figure 36: Location of tag moved 1/2 meter back from centre towards the master sensor at floor level on the area plan...........................................................................41Figure 37: Location of tag moved 1 meter back from centre towards the master sensor at floor level on the area plan...........................................................................41Figure 38: Location of tag moved 1.5 meters back from centre towards the master sensor at floor level on the area plan...........................................................................42Figure 39: Location of tag moved 2 meters form central location away from the master sensor...................................................................................................................42Figure 40: Location of tag moved 1.5 meter form central location and 1 meter to the right side.....................................................................................................................43Figure 41: Grid reference 1 m2 overlaid on the onscreen grid reference................44
5
Figure 42: A three-dimensional Cartesian coordinate system..................................45Figure 43: Tracking error................................................................................................45Figure 44: Sensor location error zone..........................................................................46Figure 45: Periodical broadcasts from the transmitting tag.......................................47Figure 46: Centre point of floor plan marked out to scale.........................................48Figure 47: Plan to scale of area in cm..........................................................................49Figure 48: Sensors in operation locating transmitting tag id: 010-000-015-099....49
6
Tables Table 1: User Use-Case Diagram Descriptions
Table 2: Advantages and disadvantages of different types of architecture
Table 3: Showing the event source, event class and event attributes
Table 4: Water test results (See Appendix A)
Table 5: Material test results (See Appendix B)
Table 6: Performance tests
Table 7: Water disturbance test results
7
Graphs Graph 1 Disturbance caused to a transmitting tags signal.
Graph 2 Disturbance caused to a transmitting tags signal.
Graph 3 The position before and after the basin full of water covers the transmitting tag.
Graph 4 Position of the transmitting tag as it is moved further away from the centre of the room. The reliability of the location detection system is effected by how far the transmitting tags are from the central position covered by the sensors.
Graph 5 Shows how the error rate becomes more acute the further the tag is moved from the central position in the room.
Graph 6: Shows the readings taken from a person wearing a transmitting tag walking from point A in the room to point B in a straight line. See Appendix C
Graph 7: Error detection results see Appendix C for the screen shots
8
AcknowledgmentsI would like to take thank project supervisor, Dr Joan Condell, for all her advice, guidance, constructive feedback support and encouragement throughout this project.
I would also like to express my thanks to Dr Kevin Curran, for providing me with the idea for this project, his guidance and support throughout the project’s development and for providing me with help and assistance throughout the semester.
I would also like to thank my family for their support and patience, friends and work colleagues for their assistance. I would like to thank the staff at Magee University for their help and assistance.
9
1 IntroductionLocation systems can be divided into two types. There are outdoor location systems (an example of which is GPS) and indoor location systems. The overall principle upon which these systems operate is to locate an object accurately. There are various types of location systems some based on wireless others based on Video imaging and digital location systems but they all have their advantages and disadvantages. Outdoor location systems for example GPS are costly to implement. The infrastructure required for these types of systems include satellites and receivers. Their performance can be affected by weather conditions and other atmospheric conditions and the accuracy which they deliver is not as accurate as that of indoor location systems.
Indoor location devices still have some problems such as the ability to locate objects exactly. This can be caused by a number of factors depending on the system being used. Each system has its advantages and its drawbacks. Some can provide a high degree of accuracy but are not suitable for manufacturing businesses as they do not perform well in these conditions due in part to interference caused by other machinery. The cost of some of these systems is also a factor as they can be very expensive to implement. Scalability is another issue that requires investigation. However in order to evaluate these systems we must look at how the different systems operate, their advantages and disadvantages. The benefits of being able to track objects accurately cannot be underestimated. Industry and business have longed to be able to effectively track objects, components, assets and people. Ultra Wide Band technologies are often described as the next generation of real time location positioning systems. In the world today industry is becoming more competitive and any technology which can provide a competitive edge is welcomed and much sought after. However not all technologies live up to their claims which can prove very costly to industry. Therefore this research project investigates the use of a real time indoor positioning system to guide robots in performing various tasks.
1.1 AimsThe aim of this project is to evaluate and test the accuracy, precision and robustness of the location detection system over a small geographical area. This will receive data that will be used to establish the locations in real time tags that transmits signals. To achieve these aims we will install an Ultra Wide Band (UWB) location detection system. This involves the wall mounting of sensors in a square grid formation at ceiling height and the connecting of sensors with cat 5 network cables to a central Ethernet switch. Timing cables are also connected to each sensor in sequence and then to a master sensor in order to achieve reliable timing results. These cables are finally connected to a computer via a network cable. The computer will have the software and licence installed to operate the hardware. The projects aim is to identify how beneficial an indoor UWB location detection system could be in the field of robotics and how industry could benefit from this new emerging technology.
10
Some Key design factors for Location systems are:
Accuracy - How often does the system meet expected requirements. Affordability - Monetary cost. Scalability - How the systems can be extended. Resource requirements - Memory, computational capability, power. Privacy - Determine location without revealing confidential information. Portability - Easily adapt and modify to future technologies. Precision - How accurate the Location system is.
1.2 ObjectivesThe objectives of this project are to investigate, demonstrate and test the UWB location detection system, the hardware performance and the software reliability, to run tests under various conditions and to determine the accuracy of locating and tracking tags using two way communications. We will test the dual radio architecture and use the feedback from the timings and signals received by the sensors to calculate the accuracy of the system for locating with the stated degree of accuracy a tag at a certain position in time and space. Testing will confirm the stated reliability of the tags under challenging environments, to ascertain if the tags can withstand harsh environmental factors that may affect their reliability compared with traditional location detection systems based on Wi-Fi.
1.3 Overview of ThesisThe approach taken to the development of this project will consist of system examining different location systems and how they are implemented with the main focus being on how well the location detection system works. A literature review will compare and detail the advances in this new emerging technology and will include detailed information about real time location systems, their reliability and how they can be successfully implemented. A review of similar applications will be carried out to assess how the development of location detection systems has been influenced by industry’s needs. Important design factors will be assessed such as precision and limiting interference. The location detection system will be analysed including the hardware and software that goes into the complete system. Developments will be discussed within the field of robotics and how they have an impact on the way we work, how they have changed industry and the workplace. Investigations into system implementation, system testing, system analysis and analysis of results will take place. This may involve areas such as Networks and Programming, Human Computer Interaction (HCI), Database Management and Wireless Technologies.
Chapter 2 will provide the Literature Review which will investigate how this new developing technology may benefit the robotics industry. Chapter 3 will provide a
11
requirement analysis detailing how a real time location system compares to similar systems. The functional and non functional system requirements will be investigated. Hardware and software requirements will also be examined. The chapter concludes with a discussion on the future of this new technology and the specific advantages and disadvantages of using real time location systems compared to similar systems. Chapter 3 will examine the problem specification and objectives of the system design processes and how these systems are used within industry. It will look at the software and also aspects of Human Computer Interaction (HCI). Story-boards screen shots will be used to portray the system in operation and how these technologies are implemented.
Chapter 4 will focus on the Robotics design and how they can make use of location detection systems. The architecture of robotics within industry will be examined in detail.
Chapter 5 will focus on the design phase of the project. The architecture of the system will be examined in detail concentrating on the hardware and software that go into making up the system. Object oriented modelling will be shown including diagrams that model the systems main processes.
Chapter 6 provides an overview of the implementation phase of the project where we will see how algorithms are used to supply a visual representation of the position of transmitting tags. Visual screen location grid display diagrams will be used with the interface showing results of tag location. The areas under discussion will be system setup and implementation, problems encountered and how these were resolved. Results will be tabularised and appraisal of these results will be summarised.
Chapter 7 will focus on performance and hardware testing of the system; different methods of testing will be investigated to ensure that the results that the system supplies are accurate, test cases of the application will be presented. Chapter 8 will provide an evaluation of the performance of the UWB real time location system - how successful the project was, did it achieve the stated requirements in terms of satisfying the projects remit? Finally, Chapter 9 will present a detailed overall summary of the project and recommendations on how future work may be influenced by this new and emerging technology.
12
2 Literature ReviewThis chapter gives a brief overview of the history of location detection systems from the earliest developments through to developments taking place today in this field. Location systems can be divided into two types which are outdoor location systems which include Sonar, VOR/DME Loran Radar and GPS. These types of location systems are characterised by reference points deployed at known positions. These types of systems use expensive infrastructure in the form of RF ground stations and satellites that are used as reference points. The majority of these systems employ RF signals to provide location information.
Indoor location systems are required to provide more accurate location detection than outdoor systems and they often have to work in harsher environments. Often RF signals are interfered with because of electromagnetic discharge from other sources. Therefore indoor location systems still have problems to overcome regarding the accuracy location detection. The preceding chapters we will be examining various systems, how they work, the benefits that they can bring and the problems that need solving. Ultra-wideband employs sonic detection methods that help overcome many of the difficulties that other indoor location systems suffer from, Ultra-wideband location detection system combines sensors and transmitting tags to provide coverage of an area referred to as intelligent space, this is where accurate information can be obtained about location of objects. The robotics industry can benefit from advanced location detection systems, using a combination of sensors and taking accurate measurements of a tagged robots location, one can accurately predicted, with better accuracy the position of a robot in relation to its surroundings. The accurate location of a robots position is important to improve reliability. Mobile robots will then be able to operate autonomously in their environment, accurate prediction of location paths and navigation direction will be achieved making them more useful productive and reliable.
2.1 Global Positioning SystemsGlobal Positioning Systems (GPS) employ 24 satellites that are in fixed orbit above the Earth and between them cover the entire Earth surface. GPS operates by sending signals that travel at the speed of light from these satellites. Each satellite has an atomic clock to ensure timing accuracy. Each satellite continually transmits messages containing information relating to the time the message was sent and the precise location of the satellite in orbit above the Earth's surface. GPS uses a combination of three satellites and Trilateration to accurately locate the position of an object. GPS is commonly used for vehicle navigation systems and are accurate to approximately 15 metres. However the drawback with GPS is that the satellite signals are not strong enough to penetrate inside most indoor environments (see Figure 1).
13
Figure 1: GPS Satellites
Trilateration (see Figure 2) is used to determine the exact position. This is similar to triangulation, but with Trilateration three reference points are needed to find a point on a 2D plane. The correct position of an object is found by calculating the intersection point of the three spheres. A fourth satellite is required to measure the time accurately. This combined with the information from the other three satellites helps locate in time and space the location of a transmitting object.
Figure 2: Trilateration. Wikipedia (2008)
In the past GPS technology was the preserve of the Military who could afford it. However recently the cost of development of GPS systems for example traffic location systems have made the technology more universally accepted. GPS technology can now be found in many appliances. For example Garmin is a company that manufactures satellite navigation devices for a wide range of customers. These include Motorists, motorcyclists, outdoors and fitness enthusiasts as well as leisure users in the aviation and marine markets (Garmin, 2008). More established tracking technologies such as Global Positioning Systems (GPS) generally do not work reliably indoors or, at best are not very accurate indoors.
14
2.2 Indoor Positioning Indoor Positioning Systems (IPS) locates and tracks objects in buildings. These may be pre-tagged objects, or discovered objects. Examples of tagged objects are patients or equipment in a hospital. Examples of discovered objects are people in burning buildings or soldiers on a battlefield. Global Positioning Systems are not suitable to establish indoor locations, since their microwaves can be scattered by roofs, walls and other objects. An IPS uses other radio technology, infrared, or ultrasound, to overcome this limitation. Infrared and ultrasound are useful in environments where wireless radio frequencies may interfere with critical equipmentThe interaction between workers, machines, tools, work areas, and the products they’re manufacturing has a significant location component. Whether in the automotive, aerospace, computers, semi-conductor, or manufacturing, Real Time Location Sensor technology can deliver improved quality at a reduced cost to make the manufacturing operations more competitive. An RTLS provides location-aware support to workers improving quality by reducing errors (and therefore reducing the cost of fixing them) and increasing efficiency by reducing process execution times. The devices and tools used by the workers can be made to know where they are and in which relation they stand to the production materials.
2.2.1 Wi-Fi (802.11) Based Indoor PositioningSound navigation and ranging is a technique that uses sound propagation (usually underwater) to navigate, communicate with or detect other vessels. There are two kinds of sonar active and passive. Sonar may be used as a means of acoustic location. Acoustic location in air was used before the introduction of radar (Wikipedia, 2008). Radio Detection and Ranging, early systems were used to detect metallic ships in dense fog but not its distance. During the Second World War Britain exploited this technology as a defence system against German aircraft attack to great effect. Radar was able to identify the range, altitude direction, or speed of both moving and fixed objects such as aircraft, ships, motor vehicles, weather formations, and terrain (Wikipedia, 2008). The Radar system works by using a transmitter to emit radio waves that are reflected back by the object that one is trying to locate. These signals are picked up by a receiver noise and interference is then cancelled out. The accuracy of the signal received can be accurately measured. Radar has many limitations. It is influenced by adverse weather conditions that can distort readings. Readings can be misinterpreted, as was the case during WW2 when the Japanese bombed of Pearl Harbour bringing America into the war. The early warning radar readings were mistaken for a flock of birds.
Wi-Fi positioning algorithms can analyse signal strength. One commercial implementation is from Ekahau. The Ekahau RTLS system uses software based location tracking to accurately track assets and people over any existing Wi-Fi network. It does this by using algorithms to compute the location of tags which have the Ekahau location protocol built in and use 2- way Wi-Fi signals to deliver the required accuracy across a geographical area, without the need to install any software or hardware in remote sites.
15
The PlaceLab architecture is an example of an indoor positioning system. It is made up of three elements. These are radio beacons in the environment: databases holding beacon location information and PlaceLab clients that estimate the location from these data (Curran, 2008). PlaceLab uses open source software developed by Intel’s research center in Seattle and an 802.11 interface which combine to predict location positions, It uses known positions that are stored in a database that are combined with 802.11 signals to allow the system to establish a users location. However when tested to find a users position in a relatively small geographical area like a University Campus, it does not achieve very good accuracy.
2.2.2 Camera Based Indoor Positioning Vicon is a digital optical and video based motion tracking system that allows for a better understanding of movement. By being able to track and analyse movement it offers solutions to real life problems. This advanced optical motion capture system consists of cameras, controlling hardware and software that is able to analyse data and output the results in a meaningful way. Vicon offers high speed, high resolution, interference-free, real-time tracking for engineering related studies. Using video and digital optical tracking techniques, (Vicon, 2008). This system is designed to be expandable and allow easy integration into the working environment. The system allows for the accurate tracking of objects by use of cameras, triangulation and software that allows users to analyse the data.
Figure 3: Vicon animation Figure 4: Vicon motion capture
A benefit of this system is that it does not use radio frequency (RF) which is sometimes prone to metallic interference. It claims to be able to provide a positional accuracy of better than 1 mm. An example of a Vicon system in operation is to create graphical representation of the way humans walk. This is achieved by attaching reflective markers to a human then using the video based motion tracking system to collect data generated by recording of the reflective markers position relative to the movement, displacement, velocity, and acceleration of selected points on the lower and upper limbs. This data is then analysed and the results can be displayed on a computer screen (see Figure 3). This system can have benefits for the robotics industry, understanding movement can advance how robots are designed and operate in a human like fashion (see Figure 4).
16
2.2.3 Ultra-Wideband location Ubisense Ultra-wideband (UWB) is a radio technology that can be used at very low energy levels for short-range high-bandwidth communications using a large portion of the radio spectrum. Ultra wideband (UWB) has gained in popularity as a new technology recently and has been the focus of much research and development. UWB offers solutions to applications, such as see-through-the-wall, security applications, family communications and supervision of children, search and rescue, medical imaging, control of home appliances, which make UWB an ideal candidate for wireless home network (Rashid, 2006). Research has shown that anything that can offer business a competitive edge is quickly adopted and usually benefits companies through improved productivity, greater efficiencies and savings. For a technology to be commercially successfully, it must address a real need (Ward, 2003).
Figure 5: Ubisense dealing with Multipath effects (Ubisense, 2008)
Therefore it is essential to investigate and identify real world needs where location awareness technology can make a difference. Ubisense provides a breakthrough in the field of accurate 3D positioning (see Figure 5). It utilises ultra- wideband in the field of Radio Frequency (RF) to deliver accurate 3D positioning which is scalable and offers real time performance. This can be used over a large geographical area, especially for in-building locations. Ubisense is targeting its sensing and middleware technologies at a number of markets, including healthcare, security, workplace productivity and military training (Steggles, 2003).
Figure 6: Ubisense compact tags, single sensors and series 7000 sensors
17
The development and deployment tools make the system easy to design, implement and maintain taking all these factors into consideration and the benefits that can be achieved using this technology. Ubisense claims that with this system, companies will benefit rapidly and have a speedy return in their initial investment. This particular commercial implementation uses proprietary tags to communicate with the series 7000 sensors (see Figure 6).
Ultra–wideband is a technology based on radio frequency which has been used by to build real time systems that can provide a high degree of positional accuracy, in real time to within 15cm. It can work in very harsh and challenging environments where conventional RFID and Wi-Fi have experienced problems with interference.
Figure 7: Car production line can be zoned using Ubisense (2008)
Examples where Ultra–wideband is used and gives good positional accuracy would be the car manufacturing industry. It has been used to identify proximity detection between objects allowing a robot to correctly and automatically select a tool required by a particular model of car without the need for human intervention or the need to reprogram the robots computer every time a car model type varies. This saves time and cost and increases efficiency as the car production lines do not have to be stopped. Within a Manufacturing site, zones can be set up using the sensors mounted on walls which can receive and evaluate low frequency signals from the location tags. Sensors are grouped together in cells which, when combined with other cells, can give coverage over a large area. Within each cell there will be a master cell that controls the other cells and allows for a high degree of positional accuracy, in real time (see Figure 7).
The comparison table below (see Figure 8) shows the different protocols used by some location detection systems. From the table UWB has a limited nominal range of 10 meters compared to Wi-Fi that has a nominal range of 100 Meters. The advantage UWB has over Wi-Fi is that it is capable of working in environments where due to interference Wi-Fi would be unsuitable.
18
Figure 8: Bluetooth WI-FI and UWB Protocols (Shahril, 2008)
2.3 Robotics and Location DeterminationThe definition of a robot is very broad. Robots will integrate methods drawn from all areas of AI, including machine learning, vision, navigation, manipulation, planning, reasoning, and speech/natural language processing. This includes industrial robot manipulators, such as those used for pick-and-place, painting, or welding operations, provide they incorporate all these three elements. Early manipulators had neither sensing nor reasoning ability; they were pre-programmed to execute specifics tasks (Bekey, 2005). Today most industrial robots are equipped with a variety of equipment that allows the robots more autonomy these include cameras, sensors and on board processors.
Today the control of many industrial processes i.e. manufacturing, production etc use automation rather than humans. Automation has brought many benefits, more choice and lower prices. Automation is used widely in the chemical, steel, paper industries and within the car manufacturing industry. Many other industries benefit from automation as it has led to increased productivity and product quality. The successful introduction of robots into the car manufacturing industries has led to increased interest into the field of robotics. Improvements have been made to control systems to make robots more adaptive and mobile. Mechanisms have been added to robots to increase their location awareness; one of these systems uses ultrasonic sensors which we will investigate in this report.
For mobile robots operating in man-made environments the 2D world assumption is usually sufficient, However, the use of ultrasonic sensors fitted to the gripper i.e. sensor-in-hand systems, of a robot-arm in order to steer the arm adaptively requires a full 3D world model (Smith, 2001).
19
Current research in autonomous robot navigation focuses on planetary exploration, mining, agricultural applications, fire applications and military applications. In the last decade we have seen many satellites sent into deep space in order to investigate planet surfaces and meteorites. These robots can operate in environments that would be hostile to humans. The information and data that these robots have gathered from the surfaces of planets has been invaluable to scientists, trying to better understand our own planets origins. There are various types of robots and the functions that they perform are as varied as they are complex. A sensing robot is one type of intelligent robot which can be defined as’ a robot which measures the surrounding environment and uses the resulting data to modify its own work pattern and adapt to the environment. (Ohba, 1992).
2.3.1 Robotic Navigation Sensors Different sensors are used to extract meaningful information and measurements that can help provide information about the robots environment. For an autonomous mobile robot to operate successfully it requires three things: perception of its environment, decision making capabilities and a way to execute information received in a meaningful manner (for example path planning and navigation to execute a goal i.e. moving from one position to another efficiently, safely and successfully avoiding objects). This requires problem solving skills and strategic planning together with navigational capabilities. The comparison table (Figure 9) shows the different types of systems currently on offer for outdoor and indoor location detection systems.
Figure 9: Wireless acoustic location with room-level resolution using ultrasound
20
Exteroceptive sensors are used to obtain information from the external environment. Robots can obtain information from their environment by using cameras that along with software allow the robot to distinguish shapes and recognise objects. Proprioceptive Sensors measure the robots internal workings, for example the battery voltage level or it can measure the rotation of the robots wheels to measure distance travelled. The advantages of mobility cannot be fully exploited without the capability of navigating (Nehmzow, 2006).
Figure 10: Beacon navigation (Lee, 2008)
Sensors can provide lots of data that can be processed to provide meaningful information. However without being able to accurately predict the robots exact location at a given time then this data received from the sensors would be unreliable and lead to errors. The goal for any location system is to be able to provide accurate positioning information within a coordinate system. This will allow a robot to move autonomously (see Figures 9 and 10).
Figure 11: Beacon aided navigation for Robots
21
Ultra wideband claims to be able to accurately locate a transmitting RF tag to within 6 cm in intelligent space. This project will investigate the accuracy of these claims and how this technology could be combined with the field of robotics’ to accurately determine in real time a robots location in relation to its environment. A robots environment can include many different geometric obstacles and terrain conditions. In order to make accurate decisions not only does a robot require sensors to detect obstacles for path planning but it also needs to know its exact location, in order to benefit from the data it receives. Using Ultra wideband we explore how this location accuracy can be achieved and how it can be applied to the field of robotics. The Robots environment representation can range from continuous geometric description to decomposition – based geometric map or even a topological map. The first step of path planning system is to transform this possible continuous environmental model into a discrete map suitable for the chosen path – planning algorithm.
Location information can also be derived from analysis of data such as video images, as in the MITSmart Rooms project. Pentland, A. (1995). Accurate object locations can be determined in this way using relatively cheap hardware, but large amounts of computer processing are required. Furthermore, current image analysis techniques can only deal with simple scenes in which extensive features are tracked, making them unsuitable for locating many objects in cluttered indoor environments.
2.7 ConclusionThis project should demonstrate that the advantages of robotics cannot be fully exploited without an effective location detection system, which will empower robots to navigate autonomously in an intelligent space therefore increasing their productivity and mobility. The Gantt chart in Appendix I outlines the project plan for semester two. The plan includes key work and milestones.
22
3 Requirements Specification and AnalysisThis chapter will outline the problem at hand. The problem will be specified in detail and an analysis of the functional and non functional requirements will be made. Also included is the list of software and hardware required for successful completion of the project. The programming languages to be used in the project are also discussed.
3.1 Problem SpecificationThe research into location detection systems has shown that there are various systems which provide location detection which include Wi-Fi, GPS and others. However this initial research has indicated that under certain environments many of these systems suffer interference, making their readings unreliable. Problems which occur are: Inaccurate location information No real time data Susceptible to interference Not expandable Not flexible Expensive to implement Relies on third party hardware
3.2 System ScenarioThe Ultra wide-band location detection system would allow automatic detection and tracking of RF tags in a large indoor area e.g. warehouse environment. Combining the tags with robotics would supply accurate location information. The information could be combined with a database to allow robots to navigate more independently with less human input therefore increasing their productivity.
3.3 Requirements AnalysisThe requirements for this project are broken down into functional and non-functional requirements. Also included is the hardware and software required to successfully complete this project.
3.3.1 Functional RequirementsFunctional requirements are statements of services the system should provide, how the system should react to particular inputs and how the system should behave in particular situations (Somerville, 2004).
Client Side1. View location of sensors on grid matrix2. Identify the sensors 3. Calibrate the sensors4. Activate the RF tags5. Calibrate the tags
23
Server Side1. The server should be a computer with optimum software and hardware.2. It will have a permanent internet connection.3. It will maintain a database of all known APs on campus.4. It will maintain location information of all clients and update this whenever
possible.5. It will be able to display this information graphically.6. It will be able to send this info to the clients.7. It will have a fully functional GUI.
3.3.2 Non-Functional RequirementsThe non-functional requirements are:
1. The system will conform to defined HCI standards.2. The system shall have a pleasing interface and shall offer a robust and
reliable service.3. The system should be flexible. 4. The connection between the client and server should be permanently
available.5. Response time should be as low as possible.6. The system should be portable.
3.4 Hardware and Software Requirements The system will require the following hardware.
1. A server to send and receive location information.2. An internet connection.3. A network of 802.11 access points.4. A laptop with a wireless NIC.5. A GPS with a serial connection for the initial stumbling (Mapping of APs)
process.
Software applications that will be required are:1. A stable platform on which to run.2. Custom made applications for client and server.
3.5 Development LanguagesA Database Management System (DBMS) allows the manipulation of data and can perform various functions including: Retrieval of records that meet a certain criteria, Cross-reference records in different tables, update records. It makes use of Datasets which can be used in conjunction with mobile devices.
Relational Database Management System contains the relational database and the software required to run SQL queries and manage users. It can also carry out restores, backups and other tasks. Some examples of Relational Database Management Systems include: Oracle, my SQL (Structured Query Language). SQL is a standard programming language in which instructions can be provided for manipulating data stored in a relational database.
24
The Ubisense System consists of a .NET application Programming interface (API) platform, written in C-Sharp, most other development is in C++. The system consists of around 100,000 lines of C-sharp code and 1 million lines of C++ code. Code based on the .NET framework can integrate on industry standards TCP/IP, HTTP and XML for web services. The .NET framework is a development and execution environment that allows languages and libraries to work together seamlessly to allow the developer to create Window based applications that are easier to build, publish and deploy. The .NET framework allows a programmer the freedom of not having to worry about the subsystems because he never has to access it directly that is because they are abstracted away from the Framework classes (see Figure 12 below).
Figure 12: .NET Framework (Woods, 2008)
3.6 ConclusionThis chapter deals with the functional and non functional requirements. It looks at various systems e.g. Ultra wide-band location detection system and lists the hardware and software requirements that required for this system to function properly. The chapter also investigates the different development languages.
25
4 System Functionality: Order of Proceedings This chapter will examine the system functionality. HCI principles will be examined, good interface design and layout. How the software that controls the location detection hardware is installed. The chapter will it will outline the problems at hand. Also included will be use case diagrams and state diagrams that will help explain some of the processes which take place.
The use case diagram in Figure 13: gives a simple bird’s eye view of the functionality of the system.
Figure 13: Use case diagram showing the functionality of the System
Use-Case Diagram are used to describe the primary actors and events and the pre and post conditions (see Table 1).
Use Case Name Load Monitoring Software
Primary Actor User
Data RF Tag ID,
Description Loads up the Application software
Preconditions Software must be installed
Post Conditions Application software loads
Use Case Name Assign TagPrimary Actor User
Data RF Tag ID,
Description The user assigns a tag to an object or person.
Preconditions The user must activate the tag.
Post Conditions The tag table is updated
Table 1: User Use-Case Diagram Descriptions
Assign Tag
Update Details
Locate Tag
Activate Tag
Load Software
26
Figure 14: State diagram for a Location Detection System
The diagram (see Figure 14) is a representation of the processes involved in a RF tag being located and tracked.
High-Level sequence diagram showing the stages and the processes required to track a RF tag (see Figure 15).
Figure 15: High-Level Sequence Diagram for Tracking RF tags.
27
4.1 HCI principles
The implementation of good HCI is very important in the design of an interface because it promotes good practice for the development of application software. By using a more a more consistent approach to software development it allows for more productivity, therefore reducing development costs (see Figure 41).
HCI principles should always be adhered to good interface design and layout. A screen should have common controls (e.g. the Help button) should be in the same position on every screen, regardless of the screens main function. Colour schemes and form layout should also be consistent. All Good HCI should aim for
Consistency Compatibility with Users’ Expectations Flexibility and Control Explicit Structure Continuous and Informative Feedback Error Prevention and Correction User Documentation and Support Visual Clarity
4.2 Interface Design
Figure 16: Good Design Principles
Well constructed web sites should adhere to the three clicks rule. This allows the user to get to the information that they require in just three clicks. The three clicks rule is important not only for web sites it should be applied to mobile devices also, as it minimises the amount of information stored in short term memory. This allows users to better retain and learn how to navigate a system.
Poor user interface causes increased mistakes in data entry and system operation, inaccessible functionality, user frustration: low productivity and/or under utilisation. Any well developed application should a clear, non cluttered design, this will minimise the time that the user takes scanning the screen to pick out relevant information required by them.
User Goals Interface
Components
User Actions Results
Functions of the System
Feedback
28
4.3 Platform Control and Location Engine Configuration
When the software for the system initially loads the user is presented the Platform Control this informs the user that connection to the Ubisense Core server has been initialized and is up and running (see Figure 17).
Figure 17: Platform Control
The main user interface shows the position of the deployed sensors in a matrix grid for configuration purposes. This will allow the user to visually reference the sensors in relation to space (see Figure 18).
Figure 18: Deployment of sensors
4.4 Sensor properties
The Sensor properties interface gives the user information relating to the location of the sensor about the X, Y and Z axis. The Yaw is the directional degrees at which the Sensor is tilted towards the Fuducial Calibration Point (see Figure 19).
29
Figure 19: Properties of the master sensor location.
Users will utilize battery powered radio tags to be mounted on a robot or worn by a person. These tags when activated will transmit RF energy to provide accurate data that allows the location of the tag to be identified through the use of triangulation. The Location engine includes software required for the sensors to track battery powered tags, in real time. Configuration wizards are used to install the software.
4.5 ConclusionThis chapter examines the system functionality. Use case diagrams give a view of the functionality of the system. Tables show use cases and state chart diagrams examine how a RF tag is located. A High-Level sequence diagram shows the stages and processes required to track a RF tag.HCI principles are examined as they contribute to good design of systems. Interface usability and good design help create systems that users feel comfortable using.The location detection system interface software is examined as well as the sensor software and how it functions..
30
Initial version
Intermediate versions
Final version
Concurrent activities
Specification
Development
Validation
Outline description
5 Design and Implementation
The aim of the project is to develop and test the accuracy of ultra wide-band location detection systems and their ability to track in real-time location and movement over a small geographical area. The hardware that will be used in this project will be supplied by the company Ubisense which specialises in location detection. The project will use battery powered radio tags and a cellular locating system to detect the location of the tags. The cellular locating system will emit precisely timed short bursts of RF energy to provide accurate triangulation of the position of the transmitting tag.
This section of the report documents the design phase of the system life cycle. UML and various diagrams are used throughout this section to visually illustrate the overall system functionality and the relationships of its many components. Firstly we consider the global design of the system.
5.1 System StructureA clearly defined workflow model which graphically depicts the system’s structures is important. The sequence of events, showing which actions take place and what data passes between the various elements, gives the developer a visual reference to aid in implementing the system (see Figure 20 below).
Figure 20: Evolutionary Development
In order to achieve a thorough and productive design phase, the many different aspects of the proposed system need to be investigated and structured. For this reason it is important to use an appropriate design methodology. While a number of different process models are available we choose to use the evolutionary based approach. With this approach a number of prototypes of the system are developed. These prototypes can be discarded if some aspect of the system is proving unworkable or can be developed further until a satisfactory system is built. Figure 11 shows a graphical representation of this approach.
31
5.2 ImplementationThe system had to be mounted at a recommended height of 3 meters above the floor level. The sensors were attached to the wall using brackets that were supplied. The sensors were then manoeuvred to point to a central location in the middle of the floor approx at a 45 degree angle ensuring a good line of sight. A spirit level was used to check that the sensors were all level before finally securing them to prevent roll. All four sensors were connected to an Ethernet switch using cat 5 network cables and one of the sensors was configured to be the Master and the other sensors slaves. A timing cable was connected from each slave to a cable port on the master sensor (see Figure 21).
Figure 21: Sensors wiring connection system
After positioning of the of the location detection system the cables were connected via a switch to a laptop. Identification of the sensors was achieved and their location established on a grid matrix. This allowed calibration of the sensors in relation to the space which they monitored (see Figure 22).
Slave Sonar Sensors Timing Cable Fuducial Calibration Point Ethernet Switch Master Sensor
Figure 22: Diagram showing the Location Detection System Setup
Laptop
Timing Cable
These lines represent the Cat 5 Network Cables from each Sensor to the Master sensor and then to the Ethernet Switch
Switch
Laptop
32
See Appendix H. for problems encountered when initially configuring and setting up the system.
5.3 DHCP SoftwareThe DHCP software required editing to see the laptop’s IP address which is used to run the Ubisense software. Laptops IP address 192.168.168.6 (see Figure 23).
Figure 23: IP Configuration settings of Laptop
After configuring the DHCP software the Ubisense Location Engine was able to pick up the IP Address of the laptop and issue dynamic IP addresses to the four sensors. Editing of the Server Software code was required (see Figure24).
. Figure 24: IP Pool address edited to range from 192.168.3.7-10.
33
Third party DHCP Server software V1.6.4 was installed, this issued a dynamic IP addresses for each sensor (see Figure 25).
Figure 25: The DHCP Server software
Verification of the IP addresses functionality was achieved by pinging each address individually to ensure that packets of data were reaching the destination host and were being returned within the local network time restraints (see Figure 26).
Figure 26: Ping request successful, confirms a working network connection
5.3.1 DHCP LogFigure 27 shows how confirmation was obtained from the DHCP log that correct connections are established over the local network and that response to requests are being received from the four client Ubisence sensors.
34
Figure 27: DHCP log in Notepad
The screen shot below shows that the DHCP server has issued the dynamic IP addresses and that they have been associated with the physical addresses of the Ubisense sensors (see Figure 28).
Figure 28: List of Dynamic addresses and the Physical addresses of sensors
5.4 Location Engine Configuration
The locating engine performs the computation of coordinates for the locations of Radio transmitting Tags in real time. These tags can be worn by people. This is achieved through multilateration, which is the process of locating the tags by accurately computing the time difference of arrival (TDOA) of a signal transmitted from the tag to the four receivers mounted at each corner of the room. Locating a Tag by measuring the TDOA of a signal transmitted from the tag and received by the synchronised sensors allows for the accurate location of the tag to be identified accurately in real time.
Location Engine Configuration in Figure 29 shows the IP addresses from 192.168.3.7-10 and the sensor status as running this indicates that all is correct in the configuration of all four Ubisence sensors. This allows for the timing synchronisation accuracy which allows each sensor to listen for the radio transmissions by the tags Timing synchronisation is achieved by the timing cables
35
that connect each sensor to the master controller sensor id no’s 00:11:CE:00:OF:56.
Figure 29: Location Engine Configuration – Sensor Status
A plan was devised of the area covered by the sensors and the identification numbers of each sensor. Master sensor ID 00:11:CE:OF:56 controls the timing for the other sensors which are connected to it (see Figure 30).
Figure 30: Location Engine Configuration – Sensor and Cells
5.5 Calibration of Sensors
The sensors are calibrated to a central fuducial point on the floor; this reference point is used to verify that the system is correctly set up. Addresses from each sensor are relayed via the master sensor to the Location platform software which is installed on the laptop. The software then constructs a grid diagram showing the location of the sensors in relation to the space which they monitor.
Each sensor had to be individually calibrated to ensure that background noise was eliminated. This is important as background noise could interfere with the accuracy of the readings (see Figure 31).
36
Figure 31: Calibration of each sensor in progress
The calibration process allowed the thresholds to be set for each sensor, this reduces interference. (see Figure 32).
Figure 32: Thresholds set for the four sensors
37
5.6 Sensors scanning area plan
Figure 33 shows results of running the software after calibration to ensure that the timing signals from the sensors were operating correctly and being able to scan the area to be covered. This was performed before introducing the Tags in order to reduce interference.
Figure 33: Sensors scanning area to be covered
As can be seen from Figure 26 above each sensor is scanning the area plan in real time searching for a Transmitting tag. The master sensor coordinates the timing signals of the scans to ensure synchronisation accuracy. When a radio tag is introduced it will transmit a signal back to the sensors located at each corner of the room. Algorithms are used to work out the difference between the timing signals and the radio transmitted signal from the tags which are picked up by the sensors. In this way the system accurately works out the location of the Tags to a degree of accuracy claimed to be within 6 cm.
Factors that may affect the accuracy of the readings could be the temperature of the room or the humidity of the room. Other factors may be reflection from objects within the room or attenuation loss from the tags. Testing of the systems performance under various conditions will allow the location accuracy claims made by the manufacturers to be examined. Movement of the tags at various speeds will be tested to see how fast the location data can be retrieved by the system and the accuracy of that data.
38
5.7 Transmitting tags
A transmitting tag is a device that is enabled with location technology, usually small enough that it can be attached to assets or carried by people (see Figure 34).
Lithium Battery
Figure 34: Internal components of a Transmitting tag
Real time location sensor tags can help locate people and assets. An example could be where one can attach tags to company laptops so that alerts can be obtained when the laptops leave the building without the authorized owner or attach tags to school projectors so that one can track which classes have the projector. Wall mounted sensors pick up the tags location as indicated by a red dot (see Figure 28).
5.8 Location sensors in operation
Results of running the system in real- time, with tag number 010-000-015-099 located at the centre of the area plan are now discussed. The location of the Tag is indicated by the red dot at the junction of the blue waves (see Figure 28).
39
Figure 35: Location of Tag at ground level at the centre of area plan
Figure 36: Location of tag moved 1/2 meter back from centre towards the master sensor at floor level on the area plan.
40
Figure 37: Location of tag moved 1 meter back from centre towards the master sensor at floor level on the area plan.
Blue wave signals become more intense and concentrated as the tag moves away from the timing sensors 00:11: CE: 00: OE: 71 and 00:11: CE: 00: OF: 38. (see Figure 37).
Figure 38: Location of tag moved 1.5 meters back from centre towards the master sensor at floor level on the area plan.
41
Tag is moved forward 2 meters from the central position and away from the master sensor.
Figure 39: Location of tag moved 2 meters form central location away from the master sensor.
Figure 40: Location of tag moved 1.5 meter form central location and 1 meter to the right side.
42
5.9 ConclusionThis chapter examines the system and how it operates. Screenshots show the location detection software in operation identifying the position and location transmitting tag id number 010-000-015-099.
6 TestingThis chapter will outline the various testing procedures that are required to ensure the functionality of the system. Testing is important as it will help locate errors that can interfere with the performance of the location detection system. Location testing was important so that results from the computer screen could be compared with the physical location of the transmitting tag; this is why the floor area accurately marked out to scale. This allowed accurate comparison of results on screen to the actual location of the tags position.
Equipment testing is important to ensure that all the components that contribute to the correct operation of the location system operate as required. Failure in any one element can have an adverse effect on the overall performance of the system. All units that go to make up an effective system require testing faulty cable can lead to system problems this which may have a knock on effect when trying to install and configure the software for the system. Making sure that each unit is working correctly at an early date in the installation can prevent many difficulties at later stages.
6.1 Location TestingLocation testing is important as it allows the user to map the actual physical location of the transmitting tag through measurements on the ground and compare these with the onscreen location of the transmitting tags. This is provided by a combination of scanning signals and timing signals from the sensors which are cross-referenced to locate the transmitting tag and display the
43
location as a grid reference point. Software is used to display the area covered by the sensors and the point where the tag is located (see Figure 41).
Figure 41: Grid reference 1 m2 overlaid on the onscreen grid reference
6.2 Calibration TestingCalibration of the sensors is required to get correct readings of the area which they cover. When this is complete the sensors will be able to pick up the precise location of the transmitting tags when they are introduced to the area covered by the sensors. The sensors locate the tags in three dimensional space from a particular point in space. The basic directions in which one can move are up/down, left/right, and forward/backward. Movement in any other direction can be expressed in terms of just these three definitions. Time is often referred to as the "fourth dimension". It is one way to measure physical change. It is perceived differently from the three spatial dimensions in that there is only one of it, and that one cannot move freely in time but can only subjectively move in one direction (see Figure 42).
Figure 42: A three-dimensional Cartesian coordinate system
6.3 Error Tracking TestingAll systems are comprised of smaller units that are required to function properly a battery in a transmitting tag that is not fully charged may not have sufficient power to transmit its location signal to the surrounding sensors and therefore supply false readings. Errors can also result from environmental factors false readings from reflective surfaces. In order to compensate for these environmental displacement errors it is important to calibrate the sensor thresholds. To do this it is important to set the “Disabled Radio” tag on the master of the cell and then
44
calibrates the four sensors. This compensates for errors in the real environment, the perpendicular distance and the angle between the sensors. The sensors measure distances with different sensitivities as the distance and the angle of transmitted tag signals varies (see Figure 43).
ø
Figure 43: Tracking error
We can measure the distance error e(d). Distance d(m) is defined as the distance from the wall mounted sensors to the transmitting tag located at a vertical distance h and the horizontal d where the position of the tag. The error rate and time calculations using of the position of the transmitting tag have shown that depending on the location of the tag the accuracy of the results can be influenced when deciding the exact location of the tag. Trilateration is employed by the wall mounted sensors to locate the position of the transmitting tag. The drawbacks of using this method are that the scanning signals from the sensors do not always intersect at one point. Environmental factors can influence the readings through reflections of signals from surface. Humidity may also cause interference.
Estimations of transmitting tag location from the results supplied by the sensors compared with the real location of the tags position in relation to a measured grid map of the area covered by the sensors indicate some discrepancies. This is especially the case the further the transmitting tags are moved away from the centre of the area being scanned. It shows that distance errors become significantly greater the further the transmitting tag is positioned from the centre of the area being covered by the sensors. There are areas beneath the sensors that are dead zones where signals from the sensors do not register (see Figure 44).
Area scanned by Sensor
Sensor
Transmitting tagLocation
Wall
Grid ref: 1 meter square
h
d
d m
45
Figure 44: Sensor location error zone
Figure 44 shows the area a wall mounted sensor scans. Testing has shown that the area directly below the sensor is a dead area and transmitting tags placed there are not picked up successfully by the sensor.
Transmitting Tag sending out periodical broadcasts of its location to the Sensors (see Figure 45).
Figure 45: Periodical broadcasts from the transmitting tag
Different types of architecture offer advantages and disadvantages some offer better privacy while others offer better accuracy. With all location detection systems there is always a trade off. One of the advantages of the Ultra wideband location detection system is it accuracy and its scalability (see Table 2).
46
Table 2: Advantages and disadvantages of different types of architecture
6.5 ConclusionThis chapter examines the system, how it operates and problems experienced with location testing to ensure the accuracy of results. Calibration testing was also examined as it is important to the overall success or failure of this project.Error testing was also examined, problems with dead zones that would not pick up a transmitting tags location.
47
7 EvaluationThis chapter explains testing that the location detection system underwent. It will summarise the test result data from testing that was carried out on the system. It will evaluate the accuracy of the ultra wideband location detection system in being able to identify and accurately locate the position of transmitting tags. The evaluation will look at how different materials may influence the accuracy of the transmitting tags location. The evaluation tests how accurately the location system can track the movement and location of a person wearing a tag from point A to point B in a room. This screen shot (Figure 46) shows the static location of the transmitting tag indicated by the red dot. The black grid represents the actual area marked out in one meter squares.
0,0 0,1 0,2 0,3 0,4 0,5 X, axis
Figure 46: Centre point of floor plan marked out to scale
The floor was mapped out in a one meter square grid. The green X indicates the Centre of the actual floor. This corresponds to Grid reference 1.5 on the Y axis
48
and 2.5 on the X axis on (See Figure 48). A number of readings were taken around the area covered by the sensors. This was done methodically starting at grid reference (0,0 Y, axis - 0,0 X, axis). Further readings were taken moving the tag to different locations were the grid lines meet finishing at grid ref reference (0,3Y, axis 0,5 X, axis). This ensured that all cross references were covered.
Figure 47 shows the grid area mapped in centimetres squared to allow for better accuracy readings to be taken. This allowed for accurate measurements to be taken and compared to the onscreen readings of the position of the transmitting tag.
Figure 47: Plan to scale of area in cm
Figure 48: Sensors in operation locating transmitting tag id: 010-000-015-099
The readings from the location of the Transmitting tags can be seen in Appendix E: Tag: 010-000-015-099 actual location was near the centre of the room grid reference (0,1.5 Y, axis - 0,2. X, axis). The location of the tag from the sensors placed it at grid reference (0,1.5 Y, axis - 0,1.8 X, axis). This shows a difference
49
on the X axis of approx (.2) of a meter difference. The Y axis reading from the sensors is almost perfect.
7.2 Evaluation Results
Performance testing evaluates the success of the location detection system claims compared with the actual results obtained through the testing of system using different scenarios. Actual results as compared with manufactures claims. Table 3 shows events that take place and the event attributes.
Event Source Event Class Event Attributes1 Active Tag A Tag Tag type: either person or equipment
user id identifies the owner of the badge2 Transmitting tag Sensor equipment is
activated .Tag identifier:x,y,z: coordinate system:uniquely identifies the badgea 3D coordinate relative to a coordinate system
3 Tag moved Sensors pick up the signal from the transmitting tag
Scanning wave signals intersect to indicate the location of the tags new location.
Table 3: Showing the event source, event class and event attributes
7.3 Evaluation results of tag location test.
Location results for Tag id 101-111-015-099
0
0.25
0.5
0.75
1
1.25
1.5
1.75
2
2.25
2.5
2.75
30.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75 3
Distance in meters X axis
Dis
tanc
e in
met
ers
Y ax
is
Graph 1: Location results for Tag id 101-111-015-999
Yellow line represents the Y axis Green line represents the X axis
Graph 1 shows the results of a tag id 101-111-015-099 being paced at grid reference .25cm on the Y axis and .25 cm on the X axis. The results are then recorded and compared with the position where the sensors indicate that the tag is. The yellow line indicates where the sensors have located the tag on the Y axis and the green line indicates were the sensors have located the tag on the X axis. The chart shows that the level of error is greater along the Y axis than along the X axis. See Appendix B for screenshots of the location data.
50
7.4 Evaluation results of water test.The purpose of the following experiments is to see how a liquid, in this case water can effect and influence the location accuracy readings of the ultra wideband location detection systems
Liquid Experiment:
Water Test Predicted Result Actual ResultMovement Difference
1 Transmitting tag id number 010-000-015-115 placed at a central location in the room under a basin filled to the top with water.
Expected there to displacement of the tags location caused by the water interfering with the signals
The location software indicates the location of the Tag id number 010-000-015-115 did move when it was placed under the basin filled with waterAfter two minutes only two of the sensors were picking up the tags location.
DiscrepanciesFrom original
position.Y axis
6cm up X axissame
2 Transmitting tag id number 010-000-015-115 placed at a central location in the room under a basin Half filled to the top with water.
Expected there to be less displacement of the tags location caused by the reduced water level interfering with the signal
Scanning wave signals were less visible at the point were the tag was located. On screen software indicated that the Tags location had moved
DiscrepanciesFrom original
position.Y axis
4cm up X axissame
3 Transmitting tag id number 010-000-015-115 placed at a central location in the room under a basin filled with 1 cm of water.
Expected there to be little interference caused by the water
All four sensors did locate the Transmitting tag. Very slight deviation from the tags original position
DiscrepanciesFrom original
position.Y axis
2cm up X axissame
4 Transmitting tag placed beneath a basin with no water in it
Expected there to be slight interference caused by the basin
All four sensors did locate the Transmitting tag. No noticeable interference caused to the position of the tag
DiscrepanciesFrom original
position.NoneY axisnoneX axis
Con
trol Transmitting tag id
number 010-000-015-115 placed at a central location in the room
All Sensors to pick up signals from transmitting tag and be able to locate the tag accurately
All sensors did pick up the transmitting tags signals.Location of tag was easily identifiable with id number visible on screen
DiscrepanciesFrom original
position.none
Table 4 Water test results (See Appendix A)
Conclusion: Results indicate that water does cause interference (See graph 1) with the location accuracy reading of the ultra wideband location detection system. The results show that the position of the tag which is being tracked does show a positional change in the presence of water. These results raise the question, if a person is wearing a transmitting tag how will their body which contains 70% fluid interfere with the accuracy readings.
51
7.5 Evaluation results of material test.The purpose of the following experiments is to see how different materials can effect and influence the operational accuracy readings of the ultra wideband location detection systems.
Material Experiment:
Material Test Predicted Result Actual ResultTag covered by
material
1 Lead Expected there to be a high degree of interference in accurately locating Tag id number 010-000-015-099 location Because of the properties that make up lead.
Interference was caused to the accurate location of tag id number 010-000-015-099 location.
High interference caused to accurately
locating the tags position
2 Tin. Expected there to be interference in accurately locating Tag id number 010-000-015-099 location
Interference was caused to the accurate location of tag id number 010-000-015-099 location. However not as much As the lead experiment
Mild interference caused to accurately
locating the tags position
3 Wood Expected there to be some interference in accurately locating Tag id number 010-000-015-099 location
No Interference was caused to the accurate location of tag id number 010-000-015-099 location
No interference caused to accurately
locating the tags position
4 Plastic Expected there to be some interference in accurately locating Tag id number 010-000-015-099 location
No Interference was caused to the accurate location of tag id number 010-000-015-099 location
No interference caused to accurately
locating the tags position
Con
trol
Tag id number 010-000-015-099 was placed on The table in the same position for all experiments. No obstacles blocking line on sight from sensors
All Sensors to pick up signals from transmitting tag and be able to locate the tag accurately
All sensors did pick up the transmitting tags signals.Location of tag was easily identifiable with id number visible on screen
DiscrepanciesFrom original
position.None
Table 5 Material test results (See Appendix B)
Conclusion: Material made lead and tin do cause interference to the accurate location of transmitting tag id number 010-000-015-099. Materials made from plastic or wood do not interfere with the capabilities of the ultra wideband location detection system from accurately receiving the signals transmitted by the transmitting tag id number 010-000-015-099. The four sensors were able to accurately locate and pinpoint the location of the tag.
52
7.6 Evaluation results of other tests carried outPerformance Test Predicted Result Actual Result Difference
1 Tag attached to an individual, who walked in a straight line from point A to Point B(See appendix C)
Sensors to pick up signals from transmitting tag and be able to locate and update the individuals movements as they walk
Sensors did pick up the transmitting tags signals as the individual walked in a straight line. Location software was updated and the individual’s progress could be monitored.
Pass
2 Tag is moved to a different position on the floor(See appendix D)
Sensor equipment is activated and locates the new position of the transmitting tag.
Scanning wave signals intersect to indicate the location of the tags new location
Pass but with some
discrepancies
3 Transmitting tag placed directly beneath sensors, at floor level
Other sensors should have picked up the signal from the transmitting tag
Sensors did not locate the Transmitting tag
Fail
4 Transmitting tag placed beneath a basin of waterScreenshot data and Table 4 (See appendix E) For more detailed results
Sensors should pick up the tags signal and location
Sensors did have difficulty picking up tags location. Introduction of water did have an effect on the results
Did locate tag with some
discrepancies
5 Position Transmitting tag at centre point of all the sensors at floor level.
Sensor location software from the four sensors scan the area to locate the transmitting tag
The software indicates the location of the Tag by a red dot on screen overlaid onto grid map of the area covered. Tag is identified by its IP address
Passed for Y axis
Error 2cm for X axis
Table 6 Other test results
Conclusion: Table 6 shows some other performance tests that were carried out on the transmitting tags and how the sensors were able to accurately locate the tags under different conditions. Performance test 1 (See appendix C) shows that the location detection software was able to successfully track and locate a person wearing a transmitting tag walking in a straight line from one location to another location.Test 2: A transmitting tag was moved to different locations within the room and readings were taken (See appendix D). There were some discrepancies with the location accuracy of the tags. Test 3: Areas directly below the sensors do not pick up the location of the transmitting tags with any degree of accuracy (see Figure 44). There appears to be a correlation between the accuracy of locating the tags and their location directly beneath the location sensors. Test 4 Disturbance was caused to the accuracy of the tags location when it was covered by water. The more water that covered the tag the greater the tags location signal was deflected. (See appendix E).Test 5 A transmitting tag was placed in the centre of the room which had been accurately mapped out (see Figure 46) This allowed an accurate reading of the tags location which could be compared to the location which the software showed was the tags location onscreen. The location detection was able to successfully pin point the tags location to 2 cm accuracy. This showed that in the centre of the
53
room the accuracy of the location detection system was more reliable here than in other locations away from the centreGraph 2: Initial disturbance caused to the transmitting signal when a full basin of water is placed over the tag id number 010-000-015-115 (see graph 1). The graph shows deflection caused to the transmitting tags signal from before the water was introduced.
A Column Graph to showing how a liquid can influence the performance of a transmitting tag.
See Water Test 1 Appendix A.
0
1
2
3
4
5
6
7
1 2 3 4 5 6 7 8 9 10
Time ( Seconds )
Sign
al m
ovem
ent i
ntef
fere
nce
(cm
)
Graph 3: Shows disturbance caused to a transmitting tags signal.
54
A Scatter Graph to showing how a liquid can influence the performance of a transmitting tag.
See Water Test 1 Appendix A
0
1
2
3
4
5
6
7
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Transmitting signal readings in (Seconds)
Sign
al m
oven
ent i
ntef
eren
ce (c
m)
Graph 4: Shows the position of the transmitting tag as it is moved further away from the centre of the room. The reliability of the location detection system is effected by how far the transmitting tags are from the central position covered by the sensors.
As water stabilises reading remains constant
Basin full of water is placed over tag
55
Graph 5: shows how the error rate becomes more acute the further the tag is moved from the central position in the room.
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
1 2 3 4 5 6 7 8 9 10Error readings as tag is moved from centre of scanned area (cm)
Erro
r Diff
eren
ce fr
om c
entre
(cm
)A Column Graph to Showing the anomolies produced as a tag
moves away from centre of the scanned area.
Graph 6: Shows the readings taken from a person wearing a transmitting tag walking from point A in the room to point B in a straight line. See Appendix C
A Scatter Graph Showing the relationship between area covered by the location detection system and the actual
location of the tags
0
0.5
1
1.5
2
2.5
3
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Length of area covered (m) For Screenshots see Appendix C
Wid
th o
f are
a co
vere
d (m
)
Tags mean Location from point a to point b
Graph 7: Shows the error detection results see Appendix C for the screen shots
56
A Column Graph Showing the Movement of a transmitting tag from point a to point b across the area
scanned by the location detection system.
0
0.5
1
1.5
2
2.5
3
1 2 3 4 5 6 7 8 9 10
Readings taken see Appendix C
Are
a co
vere
d (m
)
7.3 ConclusionThis chapter shows the evaluation results from tests that were carried out using transmitting tags and the location detection sensors to located the tags and allow readings of the results to be displayed in tabular and graphical form. The results were summarised and conclusions made depending on the results of the particular experiment.
57
8 Project conclusion
This research project is concerned with the testing of an accuracy of an indoor location system utilising ultra wideband. The project explores the location detection systems and the benefits that it can bring to the robotics industry. In robotics there are many types of navigation devices these vary from wireless systems to systems using vision. However one thing that they have in common is that without these types of location systems, robots would find it difficult to navigate autonomously. The key to robot navigation is Self localisation; this allows a robot to navigate between two points. By using a combination of map building and self localisation a robot will be able to successfully plan paths in relation to its surroundings in the real world.Determining the location of objects or people in a smart space is one of the prerequisites enabling smart applications. Ultra wideband location detection system has the advantages of being able to carry out precision localisation and handle multi-path signals. Sensor networks based ultra wideband location detection system provides greater accuracy than other systems.
Ultra wideband sonar location system is a new technology which aims to overcome many of the problems that affect other types of indoor location systems, especially in an industrial environment where interference from other mechanical and electrical devices can have an adverse effect on location detection devices. The potential for this system to deliver accurate indoor location is substantial. The system may produce beneficial side-effects for the robotics industry and other industries. The fundamental principles that govern this interaction between robot, task and environment are, at the moment, only partially understood.
This project has proposed Ultra-Wideband Location Positioning Navigation System architecture suitable for the support of mobile robotics within a typical indoor workplace environment. From my investigation and testing of the Ubisense Ultra wideband location detection system I have found that the technology is capable of locating transmitting tags while having the advantage of low power consumption with fast data transfer rates over the network which I set up and configured.
Ultra wideband location detection technology's unique characteristics make it a suitable solution for use in the tracking of robots in an indoor environment, offering acceptable positioning location although interference from obstacles in the line-of-sight of the sensors is one of the major causes of error. Line-of-sight obstacles degrade the accuracy of the time of arrival estimation of the direct path signal by making the multipath structure complex as was experienced when more than one tag was introduced to the area covered by the sensors.
58
Chapter 1 outlined the aims and objectives of this project and a brief overview of this project. Current approaches, and related research were reviewed in chapter 2, including research into various type of location detection systems GPS systems, Wi-Fi (802.11) and robotics navigation systems. It explored the benefits that this new technology may have for the robotics industry. Chapter 3 examined the requirement specifications of a Location Positioning Navigation System architecture. This chapter described the functional and non functional requirements of the proposed system. Chapter 4 examined the design and implementation of the location detection system. The architecture of system setup was examined the hard wiring of the system. Configuration of the system software and screen shots of the server software in operation was examined. An abstract view of information produced by the sensors monitoring the environment was given. Calibration of the sensors and rudimentary diagrams showing the system in operation was detailed.
Integration of the location detection system including HCI issues were examined in Chapter 5. Chapter 6 identified the testing of the location detection system; this included testing of transmitting tags within the representation of a room boundary. Estimations of transmitting tag location from results supplied by the sensors were compared with the real location of the tags position. Chapter 7 described the performance evaluation of the location detection system within the real world environment and evaluates the performance of the system against manufacturer’s claims.
This project has showed that it is possible to setup and configure a location detection system that is capable of locating transmitting tags within a small geographical area. Hodges and Louie (1994) describe an Interactive Office that gathers information about the activity of the occupants of an office; this data is generated by a number of different sensor types which monitor movement of people.
The field of robotics can also benefit from this new technology in helping to accurately detect the location of robots. However the accuracy of the transmitting tags location detection can be influenced by many factors, reflection from other surfaces, atmospheric conditions.
To summarise the evaluation of this Ultra-Wideband Location Positioning System concludes that at present the reliability for its use over a large domain with high numbers of transmitting tags may cause problems. When testing this system in a small geographical area, the performance of the system to locate a single tags position was not as accurate as claimed by the manufacturer. When multiple transmitting tags were introduced into the area covered by the sensors, the systems performance became more unreliable. This will influence the reliability and suitability of this system in the robotics field.
59
9 Potential Future Work
The area of location detection is becoming increasingly important. Knowledge of the locations of people and equipment will have a profound impact on our daily lives. Location detection systems are already having a major impact e.g. GPS systems, RFID tags etc. However current systems are ad- hoc in nature. In order for location detection systems to be really effective will require a wider deployment and a substantial user base is required. At present however this area is still largely underdeveloped. There are many limitations to be overcome such as network delays, how a distributed sensor system handles delivery of events from multiple sensors and run time errors. Future work in this area will provide a more realistic determination of its capabilities and will expose previously unconsidered application areas. It is not impossible to visualise a system will be developed to incorporate all the benefits of location detection systems which can handle thousands of objects over very large domains and can provide inter-domain functionality and the supply of multiple services.
60
ReferencesBekey, GA (2005) Autonomous Robots: From Biological Inspiration to
Implementation and Control. Cambridge, Mass.: Mit Press (Page 2)
Curran, K. Furey, E. (2008) P inpo in t ing Users w i th Loca t ion Es t imat ion Techn iques and Wi -F i Ho tspo t Techno logy Int. J. Network Mgmt (2008). Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/nem.683
Garmin, UK Press Room copyright 1996-2008 accessed online: November 2008 http://www.garmin.com/garmin/cms/site/us
GPS Explained. accessed online: November 2008
Hodges, S., Louie, G. (April 1994) Towards the Interactive Office: accessed online: March 2009 http://wwwmmd.eng.cam.ac.uk/automation/publications/papers/interactive-office.pdf
Lee, W.l, Park, JB Lee B,H (2008) Dynamic localization with hybrid trilateration for mobile robots in intelligent space. available online at: accessed December 2008.
http://www.springerlink.com/content/?k=ultra+wideband+location+ detection+systems&sortorder=asc&sw=Dynamic+localization. Published online: 23 January 2008 Springer-Verlag
Motion Capture Research: accessed online: December 2008 http://www.digital-humans.org/mocap.htm
Nehmzow, U, (2006) Scientific Methods in Mobile Robotics. Springer-Verlag London 2006
Ohba, R (1992) Intelligent Sensor Technology. Chichester, W. Sussex, Wiley. J & Sons (Page 59)
Pentland, A. (1995) Machine Understanding of Human Action.Proceedings of 7th International Forum on Frontier Telecommunication Technology, Tokyo, Japan,
Rashid, A. Khatun, S,S (2006) Performance of Ultra-Wideband Time-of-Arrival Estimation Enhanced with Synchronization Scheme.ECTI Transactions on Electrical Eng,Electronics, and Communications vol 4, No1. available online:
http://www.ecti.or.th/~explorers/EEC/0602/Content_UWB78-84.pdf
61
Shahril, M (2008) Finding the Optimal MAC Protocol for Low-Power High Data Rate Ultra-Wideband (UWB) Networks. available online at:http://www.cms.livjm.ac.uk/pgnet2008/Proceeedings/Papers/2008 038.pdf Mohd Shahril Izuan Mohd Zin, Dr Martin Hope Informatics Research Institute, The University of Salford, Salford.(2008)
Somerville, Ian (2004) Software Engineering 7th Edition, Addison-Wesley 2004
Smith, PP (2001) Active Sensors For Local Planning in Mobile Robots. Singapore: World Scientific Publishing Co. Pte. Ltd (Page 103)
Steggles, P. Batty, P. Sails, T (2003) Local positioning system: New possibilities for urban combat training. White paper, IITSEC 2003 available online:
http://www.ubisense.net/media/pdf/Local_positioning_systems-_New_possibilities_for_urban_combat_training.pdf
Ubisense Software. copyright 2007. available online: December 2008 http://www.ubisense.net/content/10.html
Ubisense Software. copyright 2007. available online at:http://www.ubisense.net/media/pdf/Ubisense_Software_LPanDev_ENV1.0.pdf
Online: Ubisense. copyright 2007. available online at: Car production line can be zoned using Ubisense
Vicon-Engineering: accessed online: December 2008 http://www.vicon.com/applications/engineering.html Vicon Polygon: accessed online: December 2008 http://www.vicon.com/_pdfs/low-res_viconpolygon-US.pdf
Ward, A. Cadman, J at Ubicomp (2003) Deploying Commercial Location-Aware Systems. White paper, Ubicomp October 2003 available online: December 2008 http://www.ubisense.net/media/pdf/Deploying_Commercial_Location- Aware_Systems_October_2003.pdf page 2
Wikipedia-Trilateration. accessed online: November 2008 http://en.wikipedia.org/wiki/Trilateration
Wikipedia. Sonar. Available at URL: accessed online: November 2008 http://en.wikipedia.org/wiki/Sonar Wikipedia. Radar. Available at URL: accessed online: December 2008 http://en.wikipedia.org/wiki/Radar
62
Wireless acoustic location with room-level resolution using ultrasound Available at URL: accessed online: March 2009http://camars.kaist.ac.kr/~hyoon/courses/cs712_2005_fall/ppt/
Woods, D, (2008) Introducing .NET Online: http://www.infm.ulst.ac.uk/~derek/Visual-Programming/Lect /index.html
63
Appendix ATest results on how a liquid can effect a tags location and detection. Prior to the test taking place a grid was marked out in one centimeter squared paper. Tag id number 010-000-015-115 was placed at the intersection of the grid points and a reading was taken as a control to identify the position of the tag.
Tag id number 010-000-015-115 uncovered is placed in a location within the room, no obstacles to prevent transmitting tag signals from reaching the sensors located at a height of 2.5meters at the corners of the room. Control:
Tags id number 010-000-015-115 identified by the small red dot on screen indicating its location in relation to the area scanned by the wall mounted sensors.
64
Test 1
Basin completely filled with water placed on top of the tag id number 010-000-015-115. The tags position indicated by the red dot has slightly moved down and to the right.This could be caused by deflection with water now covering the tag. We can see also that the tag is now only being seen by three of the sensors. Sensor 00:11:CE:71 is not locating the tag. The wave lines from the sensors are not as strong only one showing.
After 1 minute three of the sensors pick up the position of the tag id number 010-000-015-115 position. Tags id 010-000-015-115 location has moved by 6cm on the y axis from its original position. This could be due to water disturbance.
65
After leaving the Water covering the tag for 2 minute the new reading shows that the tag id number 010-000-015-115 is now only being picked up by two sensors. Tags id 010-000-015-115 location is 3cm on the y axis from its original position. This is after the disturbance of the water has had time to settle.
Test 2: Basin half filled with water
Tags id number 010-000-015-115 location placed at a central position in the room. Tag is being seen by four of the sensors and its position is identified by the red dot. Location position of tag is at the same position as the control.
Basin half filled with water placed over tag id number 010-000-015-115
66
We can see from the screen shot that the tags location has moved by 6cm upwards on the Y axis. The tag is once again only being seen by three of the sensors. Sensor 00:11:CE:71 is not locating the tag.
After leaving the tag id number 010-000-015-115 for 2 minutes the tag is only seen by three of the sensors. The displacement location of the tag has stabilised at 4cm upwards on the Y axis.
Test 3: Basin filled with 1 cm of water.
Basin filled to a depth of 1 cm with water placed over tag id number 010-000-015-115. The screen shot shows that the tags location has moved up 2 cm from its original location.
67
Test 4: Plastic basin is placed over Tags id number 010-000-015-115 no change in the location of the tag compared with the control.
68
Appendix B Test results on how different materials effect a tags location and detection.
The purpose of the following experiments is to see how different materials can effect and influence the operational accuracy readings of the ultra wideband location detection systems.
Lead Experiment:
Control
Screenshot showing the location of tag id number 010-000-015-099, Tag is placed on a table on grid paper, no obstacles preventing line of sight interference. Four sensors are each placed at a height of 2.5 meters in the four corners of the room. Sensor id numbers are 11:CE:00:0F:56 master sensor and slave sensors 00:11:CE:00:0E:, 00:11:CE:00:71and 00:11:CE:38
69
Test 1
An object made of lead is placed in front of sensors sensor 00:11:CE:00:71and 00:11:CE:00:38 one can see that the location position of the tag remains the same. There does appear to be interference caused the sensor serial number 00:11:CE:00:71 and 00:11:CE:00:38. However master sensor 11:CE:00:0F:56 and slave sensors 00:11:CE:00:0E: are able to accurately locate the position of tag id number 010-000-015-099.
70
Test 2
A piece of lead is placed in front of master sensor 11:CE:00:0F:56 and slave sensor 00:11:CE:00:0E:BF. The location of tag id number 010-000-015-099.is not picked up. The location of the lead in front of master sensor 11:CE:00:0F:56 and slave sensor 00:11:CE:00:0E:BF has caused the operation of the location detection to fail.
71
Test 3
Lead is placed on top of the tag id number 010-000-015-099. From the screenshot we can see that the tags location is not identified.
Conclusion: The lead experiment indicates that an object made of lead does interfere with the ability of the sensors to accurately locate the transmitting tag id number 010-000-015-099 location.
72
Tin Experiment:
Control: No obstacles in line of sight of sensors
Screenshot showing the location of tag id number 010-000-015-099, Tag is placed on a table on grid paper.
Test 1An object made from tin is placed in front of sensors sensor 00:11:CE:00:71and 00:11:CE:38 one can see that the location position of the tag remains the same.
73
Test 2 An object made from tin is placed in front of sensors 11:CE:00:0F:56 and sensor 00:11:CE:00:0E:BF. From the screenshot one can see that the location position of the tag is not picked up.
Test 3
Initially the tag id number 010-000-015-099 was not seen by the sensors.
74
After approx 5 sec the tag id number 010-000-015-099 was located the sensors. However sensor serial number 00:11:CE:00:71did not activate.
Conclusion: The tin experiment indicates that an object made of tin does cause some interfere with the ability of the sensors to accurately locate the transmitting tag id number 010-000-015-099 location.
Wood experimentControl: No obstacles in line of sight of sensors
Test 1An object made from wood placed in front of tag id 010-000-015-099 in order to block signals from sensor 00:11:CE:00:71and 00:11:CE:38
75
The tags position is still accurately located with no perceivable interference caused by the wood obstructing or interfering with the transmitting signals from the tag.
Test 2An object made from wood is placed in front of sensors 11:CE:00:0F:56 and sensor 00:11:CE:00:0E:BF. From the screenshot one can see that the location position of the tag remains the same.
76
Test 3
Wood is then placed on top of the tag id number 010-000-015-099 to see if a material made from wood will interfere with the detection signals from the location sensors. The screenshot shows that there is no effect on the ability of the sensors to accurately pick up the location signal from the tag id number 010-000-015-099.
Conclusion: The wood experiment indicates that an object made of wood does not interfere with the ability of the sensors to accurately locate the transmitting tag id number 010-000-015-099 location.
77
Plastic Experiment:Control: No obstacles in line of sight of sensors
Test 1Plastic is placed in front of tag id number 010-000-015-099
An object made from plastic is placed in front of sensors sensor 00:11:CE:00:71and 00:11:CE:38 The location position of the tag remains the same.
78
Test 2
An object made from plastic is placed in front of sensors 11:CE:00:0F:56 and sensor 00:11:CE:00:0E:BF. From the screenshot one can see that the location position of the tag remains the same.
Test 3 Plastic is then placed on top of the tag id number 010-000-015-099 to see if a material made from plastic will interfere with the detection signals from the location sensors. There has been no effect on the ability of the sensors to accurately pick up the location signal from the tag id number 010-000-015-099.
79
Conclusion: The plastic experiment indicates that an object made of plastic does not interfere with the ability of the sensors to accurately locate the transmitting tag id number 010-000-015-099 location.
80
Appendix CResults from tracking a person wearing a tag from point A to point B.
Test to track and time the movement of a person wearing a tag id number 010-000-015-115 moving in a straight line from location A to location B across the area covered by the four sensors.
Test Data for walking in a straight line from location A to location B
Reading 1
Reading 2
81
Reading 3
Reading 4
82
Reading 5
Reading 6
83
Reading 7
Reading 8
84
Reading 9
Reading 10
Conclusion: Ten readings were taken at points along the path from location A to location B. The resulting data show that the system was able to locate and track the path of the person wearing the tag id number 010-000-015-115 as they transverse the area covered by the four sensors
85
Appendix D Results of tests carried out on the accuracy readings of a tags location.Location accuracy of Tag 010.000.015.099
Tag 010.000.015.099 located in a diagonal 25 cm from bottom right hand wall (X) and 25 cm from opposite wall (Y).
Tag 010.000.015.099 located in a diagonal 50 cm from bottom right hand wall (X) and 50 cm from opposite wall (Y).
y
X
y
X
86
Tag 010.000.015.099 located in a diagonal 75 cm from bottom right hand wall (X) and 75 cm from opposite wall (Y).
Tag 010.000.015.099 located in a diagonal 1 m from bottom right hand wall (X) and 1 m from opposite wall (Y).
X
X
X
y
y
87
Tag 010.000.015.099 located in a diagonal 1.25 m from bottom right hand wall (X) and 1.25 m from opposite wall (Y).
Tag 010.000.015.099 located in a diagonal 1.5 m from bottom right hand wall (X) and 1.5 m from opposite wall (Y).
X
y
X
y
88
Tag 010.000.015.099 located in a diagonal 1.75 m from bottom right hand wall (X) and 1.75 m from opposite wall (Y).
Tag 010.000.015.099 located in a diagonal 2 m from bottom right hand wall (X) and 2 m from opposite wall (Y).
Tag 010.000.015.099 located in a diagonal 2.25 m from bottom right hand wall (X) and 2.25 m from opposite wall (Y).
X
y
X
y
89
Tag 010.000.015.099 located in a diagonal 2.5 m from bottom right hand wall (X) and 2.5 m from opposite wall (Y).
Tag 010.000.015.099 located in a diagonal 2.75 m from bottom right hand wall (X) and 2.75 m from opposite wall (Y).
X
y
X
y
90
Tag 010.000.015.099 located in a diagonal 3 m from bottom right hand wall (X) and 3 m from opposite wall (Y).
Tag 010.000.015.099 located in a diagonal 3.25 m from bottom right hand wall (X) and 3.25 m from opposite wall (Y).
X
y
X
y
91
Tag 010.000.015.099 located in a diagonal 3.5 m from bottom right hand wall (X) and 3.5 m from opposite wall (Y).
X
y
92
Appendix E Test results: Liquid disturbance test Liquid test: How does the Tag id 010.000.015.099 operate when covered by a liquid i.e. (water).
Figure shows location of tag id 010.000.015.099 on table before being covered by a basin of water.
93
On placing the basin of water over the tag id 010.000.015.099, the location indicated by the red dot appears to move backwards slightly.
On leaving the basin to settle the tag id 010.000.015.099 location has moved almost back to its original position.
94
Appendix F Test Results: Three tags transmitting at the one time.
If three tags are brought into the area being scanned it causes confusion. The sensors rapidly skip from one Tags location to the next one. Tag id 010.000.015.248 Tag id 010.000.015.199 and Tag id 010.000.015.115.
Tag id 010.000.015.248 location
Tag id 010.000.015.199 location
95
Tag id 010.000.015.115. location
96
Appendix GConfiguration and setup of the location detection system
Ubisense Location detection system.
Status LEDs Sensors
Eight port Netgear Switch used to connect Sensors to computer.
Transmitting tagsComputer setup and how the location detection system is setup
97
Appendix HProblems with the setup and configuration of the location detection system
Third party Server software used to issue Dynamic IP addresses to the Ubisense sensors in order for them to function properly. When using third party software it is sometimes difficult to find support and documentation when problems occur.
When initially running the Server software a conflict occurred with the IP addresses not being recognised by Ubisense Location Engine. This problem was identified by the Led codes from the sensors see figure: below. The code combination pointed to a DHCP timeout.
Figure: status codes indicated by LEDs on the Ubisense sensors
98
Figure: Location Engine Configuration showing initial allocation of IP addresses form 192.168.3.10-14 by the DHCP Server software. The status of the IP Addresses is unknown.
Below confirmation of the DHCP timeout, use of the ping command verified that there was a problem with the addresses issued by the DHCP Server.
Figure: Failure of packets sent to Dynamic IP address 192.168.3.14 allocated by the DHCP Server application.
99
Appendix I
PROJECT SCHEDULE January February March AprilRESEARCH ACTIVITES W1 W2 W3 W4 W1 W2 W3 W4 W1 W2 W3 W4 W1 W2 W3 W4INVESTIGATIVE/SELECTIVE Further location determination based research on systems Learn configuration of Ubisense hardware and software Learn the necessary software packages for algorithms Set up necessary experimental framework for evaluation. SYSTEM IMPLEMENTATION Create experiments
Enhance prototype
Evaluate results Compare system with Vicon camera based system Build in slack for unplanned problems INTEGRATION AND TESTING WRITE UP OF THESIS
100
