Upload
vocong
View
225
Download
0
Embed Size (px)
Citation preview
Robotics and Autonomous SystemsLecture 1: Introduction
Simon Parsons
Department of Computer ScienceUniversity of Liverpool
1 / 41
Acknowledgements
• The robotics slides are heavily based on those that RolandSiegwart and Illah Nourbakhsh and provide along with theirbook:
• The agents slides are heavily based on those given to meby Mike Wooldridge and are taken from his book:
• The module information and much of the material onLeJOS was provided by Davide Grossi.
• Naturally, all errors are my responsibility.
2 / 41
Today
• Module Information• Aims and learning outcomes• Contents, outline• References• Practical matters, assessment
• Introduction• What is robotics?• What are agent systems?• Some examples of robots.
3 / 41
What is a robot?
• But first, what do you think?
4 / 41
Module aims
1 To introduce the student to the concept of an autonomousagent.
2 To introduce the key approaches developed fordecision-making in autonomous systems.
3 To introduce a contemporary platform for programmingagents and multiagent systems.
4 To introduce the key issues surrounding the developmentof autonomous robots.
5 To introduce a contemporary platform for experimentalrobotics.
5 / 41
Learning outcomes
1 Explain the notion of an agent, how agents are distinctfrom other software paradigms (e.g., objects), and judgethe characteristics of applications that lend themselves toan agent-oriented solution
2 Identify the key issues associated with constructing agentscapable of intelligent autonomous action
3 Describe the main approaches taken to develop suchagents
4 Use a contemporary agent programming platform (e.g.AgentSpeak) for developing significant software orhardware-based agents
5 Identify key issues involved in building agents that mustsense and act within the physical world
6 Program and deploy autonomous robots for specific tasks
6 / 41
. . . and the following soft skills
1 You will be able to practice how to work in groups:• discussing solutions together• distributing tasks and managing time• giving and keeping deadlines• respecting each others ideas
2 You will be able to practice how to manage a computerscience project spanning over several weeks
• planning ahead• keeping track of design challenges and choices made
Don’t underestimate the challenge of either of theseaspects.
7 / 41
Module structure
• PART 1: RoboticsPrinciples of robotics, the NXT platform and the LeJOSprogramming language.
• PART II: Autonomous systemsPrinciples of agent theory, robots viewed as agents(autonomous systems), and the Jason programminglanguage, agent coordination.
8 / 41
Module outline
• 10 weeks (of 12 weeks)• 30 lectures with a mix of Theory (principles of robotics and
agent systems) and Practice (code)• 10 supervised lab sessions (1 per week) + free lab access• Self study and practice (as much as needed to understand
the material and gain mastery of the subject)• Two assignments (combined weight 85%) to be carried out
in teamsNote: This is the main assessment for the module.
• The first assignment will be published early in Week 2• Teams are decided by the lecturer and are non-negotiable• The individual mark for each assignment will depend on
the mark obtained by the team, wighted by the amount ofcontribution of the individual to the team
9 / 41
Assessment
• First assignment (40%)• Second assignment (45%)• Homework (10%)• In-class quiz (5%)
10 / 41
Peer group assessment
11 / 41
Project log
• Each group will be asked to keep a project log, madeavailable online
• The project log should contain information about, forinstance:
• which subproblems is the group tackling• what solutions have been considered and how they are
performing• what problems is the group facing• . . .
• It should, at the end, give an idea of the trajectory thegroup has taken towards the proposed solutions of theassignments
• I will use it to monitor how each group does and it willcontribute to the mark for each assignment.
12 / 41
Practical matters
• Access to robots and lab:• The Robot Lab will be open during the usual lab hours.• You can borrow robots during Helpdesk opening hours.• You will leave your student card as a deposit when you
check out a robot.• Module website:
• cgi.csc.liv.ac.uk/˜sp/teaching/comp329-2014/
• It will contain slides, bibliography, assignments and extramaterials and will be used for public announcementsconcerning the course
13 / 41
Bibliography
Robotics
Agent Systems
14 / 41
What is a robot?
15 / 41
What is a robot?
• “. . . a programmable, multifunction manipulator designed tomove material, parts, tools, or specialized devices throughvariable programmed motions for the performance of avariety of tasks” Robot Institute of America (1980)
• “. . . [a] physical agent that performs tasks by manipulatingthe physical world” Russell and Norvig (2003).
16 / 41
What is a robot?
• The word “robot” was first used in Karel Capek’s play“Rossum’s Universal Robots” in 1920
17 / 41
What is a robot?
• Isaac Asimov coinedthe term “robotics” in1942.
18 / 41
What is an agent?
sensors
effectors
percepts
actions
Environment
• An agent is a system that is situated in some environment,and is capable of autonomous action in this environment inorder to meet its delegated objectives.
19 / 41
What is an agent?
sensors
effectors
percepts
actions
Environment
• The fundamental question is what action(s) to take for agiven state of the environment.
20 / 41
What is mobile robotics?
• In mobile robotics this becomes three questions:
?
• Where am I ?• Where am I going ?• How do I get there ?
• The robotics part of this course is about answering thosequestions.
21 / 41
Applications
• The following are some deployed robots.• From the more mundane to the more experimental.
22 / 41
Automatic Guided Vehicle
• Used to transport motor blocks from one assembly stationto an other.
23 / 41
Automatic Guided Vehicle
• Guided by an electrical wire installed in the floor but it isalso able to leave the wire to avoid obstacles.
24 / 41
Helpmate
Used in hospitals for secure trans-portation of items such as medica-tion, samples (medical records, x-ray film).
• Not clear if the company (Pyxis) is still in business.
25 / 41
Another hospital delivery robot
• Experimental delivery robot evaluated by healthcarecompany Kaiser Permanente.
26 / 41
BR700 cleaning robot
Sold by Karcher Inc.
The navigation systemis based on a sophis-ticated sonar systemand a gyro.
• Doesn’t seem to be in production any more.
27 / 41
iRobot Roomba
Successful robotic vacuumcleaner
6 million sold by mid 2011
Simple, but effective, controlstrategy
Touch sensors
Sonar in the more advancedmodels
28 / 41
Pioneer inspection robot
• Teleoperated robot used to inspect the sarcophagus atChernobyl.
29 / 41
Tiburon ROV
• Teleoperated robot used by the National Oceanic andAtmospheric Administration for underwater exploration
30 / 41
Sojourner
• Sojouner was deployed on Mars.
• On Mars, teleoperation is a bit of a handicap. Why?
31 / 41
Mars Yard
32 / 41
iRobot Packbot
• A rugged, teleoperated platform for the military
33 / 41
Grey Eagle
• A UAV much used by the US military.
34 / 41
Pulstech Forester
• First “industrial like” walking robot. Leg coordination isautomated, but human operator navigates.
35 / 41
Aibo entertainment robot
36 / 41
DARPA Grand Challenge
• The Grand Challenge involved autonomous navigationthrough the desert.
37 / 41
DARPA Urban challenge
• The Urban Challenge took this into traffic.
38 / 41
Google Car
• The Google car is intended to make this technologymainstream.
39 / 41
Unmanned vehicles
• Expect to see a range of other autonomous vehicles.
40 / 41
Summary
• This lecture gave you the necessary information on themodule.
• It also looked at some basic ideas in the area of mobilerobotics.
• We identified the main problems to be solved.• The course will cover these topics (and others).
• We also looked at some deployed robots.
41 / 41