Upload
andrew-vallejo-sanderson
View
613
Download
1
Embed Size (px)
Citation preview
||Autonomous Systems Lab
Roland Siegwart, ETH Zurichwww.asl.ethz.chwww.wysszurich.ch
Zurich, June 21, 201621.06.2016Roland Siegwart 1
Status quo and Trends of Robotics
Shaping the future
||Autonomous Systems Lab
� Introduction | Digitalization and Recent Direction in Robotics
� Robot Systems | Examples from ASL/RSL at ETH Zurich
� Mobile Robot Navigation | from Perception to Autonomous Navigation
� Application Fields and Conclusion
21.06.2016Roland Siegwart 2
Content
||Autonomous Systems Lab
� Microelectronics / Computing
� Internet / Information
� Mobile Devices / Connectivity
� Robotics / Intelligent Systems
21.06.2016Roland Siegwart 6
Technologies Disrupting Services | digitalization / industry 4.0
||Autonomous Systems Lab
� Microelectronics / Computing
� Internet / Information
� Mobile Devices / Connectivity
� Robotics / Intelligent Systems
21.06.2016Roland Siegwart 7
Technologies Disrupting Services | digitalization / industry 4.0
Physically Interacting Machines
→ exponential increase of complexity
||Autonomous Systems Lab 21.06.2016Roland Siegwart 8
Robotics today (Tesla) https://www.youtube.com/watch?v=8_lfxPI5ObM
||Autonomous Systems Lab 21.06.2016Roland Siegwart 10
Next generation of Robots| mobile, smart, connected, adaptive and closer to humans
Cyborgs
Service and Personal Robots
Industrial Robots
||Autonomous Systems Lab
� Complex machines with high added value
21.06.2016Roland Siegwart 12
Fascinating Robotics
DARPA Robotics Challenge07.06.2015,
Team NEDO-JSK, Japan12 x original speed!!
Spot | hydraulic quadruped
OceanOne | Humanoid Underwater Robot
https://www.youtube.com/watch?v=M8YjvHYbZ9w
https://www.youtube.com/watch?v=8P9geWwi9e0
http
s://w
ww.
yout
ube.
com
/wat
ch?v
=p1H
mgP
9l4V
Y
||Autonomous Systems Lab 21.06.2016Roland Siegwart 16
“Soft Robots” | torque / force controlled robots
lightweight robot
BaxterYuMi
ANYbotics
||Autonomous Systems Lab 21.06.2016Roland Siegwart 18
Mobile Platforms
ANYbotics
||Autonomous Systems Lab
� Mission and Dedication� To create intelligent robots and systems that operate
autonomously in complex and dynamic environments.
� Research Focus� Novel robot concepts that are best adapted for ground, air, or
water based applications.� New algorithms for perception, localization, abstraction, mapping,
and path planning that will enable autonomous operation in challenging environments.
21.06.2016Roland Siegwart 24
Autonomous Systems LabInstitute of Robotics and Intelligent SystemsProf. Dr. Roland Siegwart
||Autonomous Systems Lab
Design of Rolling, Swimming, Walking and Flying Robots
21.06.2016Roland Siegwart 26
||Autonomous Systems Lab
rezero (2010)| the ball balancing robothttps://www.youtube.com/watch?v=ACohrH64YKs
BeachBot (2014, with Disney)| the beach artisthttps://www.youtube.com/watch?v=eBRrQBPtdak
Vertigo (2015 with Disney)| the ultimate wall climberhttps://www.youtube.com/watch?v=KRYT2kYbgo4
Scalevo (2015)| the stair-climbing wheelchairhttps://www.youtube.com/watch?v=3lb_8nmy90c
Ultimate Rolling Robots – designed by students
Roland Siegwart 21.06.2016 32
||Autonomous Systems Lab
Quadruped Legged Locomotion
21.06.2016Roland Siegwart 40
ANYbotics
||Autonomous Systems Lab
� http://www.youtube.com/watch?v=Jql6TSyudFE
42
Efficient Walking and Running | what nature evolved (Extreme Jumpy Dog)
21.06.2016Roland Siegwart
||Autonomous Systems Lab Roland Siegwart 43
Efficient Walking and Running | serial elastic actuation
21.06.2016
https://www.youtube.com/watch?v=6igNZiVtbxU
||Autonomous Systems Lab
� High mobility � “to go where today only humans can go”
� 10 kg of payload� 2 h of continuous operations
21.06.2016Roland Siegwart 48
ANYmalan electrically actuated dog for real-world scenarios Prof. Marco Hutter
https://www.youtube.com/watch?v=EI1zBTYpXW0
||Autonomous Systems Lab
� Helicopters: � < 20 minutes� Highly dynamic and agility
� Fixed Wing Airplanes: � > some hours; continuous flights possible� Non-holonomic constraints
� Blimp: lighter-than-air� > some hours (dependent on wind conditions);� Sensitive to wind� Large size (dependent on payload)
� Flapping wings� < 20 minutes; gliding mode possible� Non-holonomic constraints� Very complex mechanics
50
UAV (Unmanned Aerial Vehicles) | flight concepts
21.06.2016Roland SiegwartFesto BionicOpter
||Autonomous Systems Lab
Reely (2009 – with Disney)| the flying reel
https://www.youtube.com/watch?v=RF6OyKKmrX8
Skye (2012 – with Disney)| the omnidirectional blimp
https://www.youtube.com/watch?v=qXvl3anK3w0
PacFlyer/wingtra (2013)| the VTOL UAV
https://www.youtube.com/watch?v=QADvPDWtgFU
Flying Robots | new ways of flying
Roland Siegwart 21.06.2016 52
||Autonomous Systems Lab
� Based on Mass & Power Balance� Need for precise scaling laws
(mass models)
Roland Siegwart 56
Solar Airplane |design methodology for continuous flights
21.06.2016
Airplane Parts• Solar cells• Battery• Airframe• …
Î Total massAerodynamic & ConditionsÎ Power for level Flight
||Autonomous Systems Lab
Skysailor (2008)| pioneering continuous flights| 3.2 m, 2.3 kg
https://www.youtube.com/watch?v=IU4BoEFOEKI
(2012)| robust and versatile solar plane| 3 m, 3.8 kg
(2015)| 81 hours non-stop in summer 2015| 5.64 m, 6.2 kg
https://www.youtube.com/watch?v=8m4_NpTQn0E
Flying Robots – fixed wing
Roland Siegwart 21.06.2016 58
||Autonomous Systems Lab
Mobile Robot Navigation
21.06.2016Roland Siegwart 59
?
||Autonomous Systems Lab 21.06.2016Roland Siegwart 63
“Seeing” | Laser-based 3D mapping
||Autonomous Systems Lab 21.06.2016Roland Siegwart 64
“Seeing” | Visual-Inertial Motion Estimation
https://www.youtube.com/watch?v=yvgPrZNp4So
||Autonomous Systems Lab 21.06.2016Roland Siegwart 65
“Intelligent Smartphone” | perceiving and understanding the environment
http
s://w
ww.
yout
ube.
com
/wat
ch?v
=yvg
PrZ
Np4
So
phab2proLenovo
||Autonomous Systems Lab
� Humans are unbeatable in taking decisions in complex situations
� Technology is better in simple but fast decisions
(ABS, ESP, )
“Understanding” the world
21.06.2016Roland Siegwart 72
||Autonomous Systems Lab
Laser-based navigation in complex environments
3D mapping and path planning
21.06.2016Roland Siegwart 77
||Autonomous Systems Lab
� In collaboration with � University of Freiburg,� Univ. of Oxford� KU Leuven� RWTH Aachen � BlueBotics
21.06.2016Roland Siegwart 97
Autonomous navigation in citiesEUROPA - European Robotic Pedestrian Assistant
https://www.youtube.com/watch?v=HE0NK7rFfpk
||Autonomous Systems Lab
Real-time on-board Visual-Inertial Navigation
21.06.2016Roland Siegwart 98
||Autonomous Systems Lab 21.06.2016Roland Siegwart 101
OKVIS | Vision-Only vs. Visual-Inertial in Optimization
IMU termsReprojection errors (weighted)Cost
www.skybotix.com
i: camera index; k: camera frame index; j: landmark index.
||Autonomous Systems Lab 21.06.2016Roland Siegwart 130
V-Charge | using close-to-market sensors
cameras
Wheel encoders
ultrasound
11.11.2015
||Autonomous Systems Lab
V-Charge | Vision and Results
21.06.2016Roland Siegwart 134
https://www.youtube.com/watch?v=7xQfKTAtyNU
||Autonomous Systems Lab
Dri
ving
Spe
ed
Complexity of Environment → Perception and Interaction 21.06.2016Roland Siegwart 137
Autonomous Cars | roadmap
Autonomouscar freeway
Autonomouscar urban
Fully autonomousCar
(you can sleep)
||Autonomous Systems Lab 21.06.2016Roland Siegwart 142
Autonomous Cars | a bright future without traffic jams
||Autonomous Systems Lab
� Swarm of small helicopters� Vision-inertial navigation (one camera and IMU, GPS denied)� Fully autonomous with on-board computing� Feature-based visual SLAM� robust against lighting changes and large scale changes
149
UAV | Vision only navigation
Proto 1
Proto 2
Proto 3
www.sfly.ethz.ch/
21.06.2016Roland Siegwart
http
s://w
ww
.you
tube
.com
/wat
ch?v
=vH
pw8z
c7-J
Q
||Autonomous Systems Lab 150
UAV | collision avoidance and path planning
Proto 1
Proto 2
Proto 3
21.06.2016Roland Siegwart
� Real time 3D mapping (on-board)� optimal path planning considering localization uncertainties
https://www.youtube.com/watch?v=95XGvEs9iTs
http
s://w
ww
.you
tube
.com
/wat
ch?v
=95X
GvE
s9iT
s
http
s://w
ww
.you
tube
.com
/wat
ch?v
=D6u
Vejy
Mea
4
http
s://w
ww
.you
tube
.com
/wat
ch?v
=-cm
-HkT
I8vw
||Autonomous Systems Lab 21.06.2016Roland Siegwart 158
Collaborative Visual-Inertial Navigationin collaboration with
https://www.youtube.com/watch?v=9PprNdIKRaw
Prof. Marco Hutter
http
s://w
ww
.you
tube
.com
/wat
ch?v
=9P
prN
dIK
Raw
||Autonomous Systems Lab
Applications
21.06.2016Roland Siegwart 159
||Autonomous Systems Lab
StaticStructured, 2D
StaticUnstructured, 3D
DynamicStructured, 2D
StaticDynamicSemantics
DynamicUnstructured, 3DAc
tion
s–
from
sim
ple
mot
ion
to c
ompl
ex i
nter
acti
on
Environment - from static 2D grid maps to 3D cognitive maps
AdvancedInteraction
MobileManipulation
TactileManipulation
AutonomousNavigation
Complexityof Services
Robotics Roadmap
Autonomouscar urban
Tour-Guides
Householdassistant
AGVs
All-terrainnavigation
Autonomouscar freeway
Industrialservices
Toys
Householduniversal
dynamic
Search and Rescue
Agriculturerobots
Roland Siegwart 16021.06.2016
Constructionrobotics
||Autonomous Systems Lab
� Industrial transportation� Cleaning� Medical robotics� Entertainment / edutainment
� Logistics� Autonomous Cars
� Industrial inspection� Surveillance and rescue� Construction and mining� Agriculture
� Health and elderly care� Personal / services robots
161
Opportunities / Markets
The coffee servantNesspresso / Bluebotics, Switzerland
21.06.2016Roland Siegwart
YuMi
||Autonomous Systems Lab
� $120 Mio donation from HJ Wyss
� 6-7 years� Focus� Robotics� Regenerative Medicine
� 10 technology transfer projects running, more in the pipeline
21.06.2016Roland Siegwart 164
Bridging the “valley of death” | The Wyss Zurich
www.wysszurich.ch
||Autonomous Systems Lab
Zurich Area | a melting pot for robotics technology� Initiatives / Spin-offs / Industrial Collaborations
21.06.2016 Roland Siegwart168
Center for Learning Systems
||Autonomous Systems Lab
� Chris Anderson� CEO of 3DRobotics� WIRED, editor-in-chief until 2012
21.06.2016Roland Siegwart 172
Switzerland, the Silicon Valley of Robotics
||Autonomous Systems Lab
� Robotics is a very fascinating engineering field� However, robotics is a very very hard problem� Design and precision mechanics� Perception� Physical interaction� Intelligence
� The way forward� A single fine-tuned demonstration is not enough� Hype / Bobble� Overselling will bounce back
� However, there are low hanging fruits, especially for Switzerland
21.06.2016Roland Siegwart 173
Conclusion
AdvancedInteraction
MobileManipulation
TactileManipulation
AutonomousNavigation
||Autonomous Systems Lab 21.06.2016Roland Siegwart 178
ASL Team