Shaping the future Robots Leaving the Production Halls Opportunities and Challenges Prof. Dr. Roland Siegwart www.asl.ethz.ch www.wysszurich.ch APAC INNOVATION SUMMIT 17 Hong Kong Science Park Science, June 5, 2017 Roland Siegwart 15.06.2017 1
Technologies disrupting services digitalization / industry 4.0 1981 2017 Personal Computer Internet information sharing Smart Phones permanent connectivity IoT / Cyber-Physical Systems / Robots Data World is connecting with physically exponential increase of complexity Roland Siegwart 15.06.2017 2
Robotics today Industrial Robots (Changan-Ford China ) Roland Siegwart 15.06.2017 3 248 000 industrial robots sold in 2015 68 000 in China https://www.youtube.com/watch?v=seioqy0oxji
Robotics tomorrow Service Robots outside the production halls Roland Siegwart 15.06.2017 4 SpotMini electic quadruped, Boston Dynamics https://www.youtube.com/watch?v=tf7ievtdjng https://www.youtube.com/watch?v=8p9gewwi9e0
Service Robots Key Challenges Robots that can dealing with uncertain and partially available information Robots that see, feel and understand their environment Robots with torque and force control for tactile interaction ( soft robots ) 50x speed https://www.youtube.com/watch?v=gy5g33s0gzo Robots with intuitive human-machine interfaces Robots that learn and adapt every day Roland Siegwart 15.06.2017 5
@ ETH Institute of Robotics and Intelligent Systems Prof. Dr. Roland Siegwart Mission and Dedication To create intelligent robots and systems that operate autonomously in complex and dynamic environments. Research Focus Body Design Brain Navigation 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. Roland Siegwart 15.06.2017 6
Service Robots designed for challenging tasks BeachBot (with Disney) developed by students the beach artist https://www.youtube.com/watch?v=ebrrqbptdak Vertigo developed by students the ultimate wall climber https://www.youtube.com/watch?v=kryt2kybgo4 Roland Siegwart 15.06.2017 7
Service Robots designed for challenging tasks wingtra developed by students the VTOL UAV https://www.youtube.com/watch?v=qadvpdwtgfu 81 hours non-stop in summer 2015 5.64 m, 6.2 kg https://www.youtube.com/watch?v=8m4_nptqn0e Roland Siegwart 15.06.2017 8
Service Robots designed for challenging tasks Scewo developed by students the stair-climbing wheelchair https://www.youtube.com/watch?v=3lb_8nmy90c AnyMal soft interaction with the environment the ultimate quadruped https://www.youtube.com/watch?v=ei1zbtypxw0 Roland Siegwart 15.06.2017 9
Seeing Laser-based 3D mapping Mapping & Localization scan t 1 scan matching scan t 2 Zukunft der Robotik 24.11.2016 10
Seeing Visual-Inertial Motion Estimation Image 1 Image 2 https://www.youtube.com/watch?v=yvgprznp4so Zukunft der Robotik 24.11.2016 11
OKVIS Open Keyframe-based Visual-Inertial SLAM (tight coupling of vision and IMU) Cost Reprojection errors (weighted) IMU terms i: camera index; k: camera frame index; j: landmark index. Roland Siegwart 15.06.2017 12
OKVIS in Action Roland Siegwart 15.06.2017 13
ROVIO Robust Visual Inertial Odometry robo-centric representation EKF based IMU-Vision fuses projected intensity errors (instead of reprojection errors) Procedure feature detection & image patch is extracted Derivation of an intensity based error terms dimension reduction of error term by QR-decomposition directly used as Kalman filter innovation Roland Siegwart 15.06.2017 14
ROVIO Robust Visual Inertial Odometry https://www.youtube.com/watch?v=zmaisvy-6ao&list=pljol3sa8g75rnj0valyl0bbftnuhwwe1g&index=2 [M. Bloesch et al (2015). Robust Visual Inertial Odometry Using a Direct EKF-Based Approach, IROS] Roland Siegwart 15.06.2017 15
Autonomous Flying Robots Visual-Inertial Sensor Combined High- and Low-level Control and processing power Robotic arm Versatile algorithm deployment Motion Estimation SLAM Dense 3D reconstruction Path-planning algorithms Obstacle Avoidance Roland Siegwart 15.06.2017 16
UAV vision only navigation Vision-inertial navigation (one camera and IMU, GPS denied) Fully autonomous with on-board computing Scale estimation Feature-based visual SLAM robust against lighting changes and large scale changes www.sfly.ethz.ch/ https://www.youtube.com/watch?v=vhpw8zc7-jq Proto 2 Proto 1 Proto 3 Roland Siegwart 15.06.2017 17
UAV collision avoidance and path planning Real time 3D mapping (on-board) optimal path planning considering localization uncertainties Roland Siegwart 15.06.2017 18 https://www.youtube.com/watch?v=95xgves9its https://www.youtube.com/watch?v=d6uvejymea4 https://www.youtube.com/watch?v=-cm-hkti8vw
Collaborative Visual-Inertial Navigation in collaboration with Prof. Marco Hutter https://www.youtube.com/watch?v=9pprndikraw https://www.youtube.com/watch?v=9pprndikraw Roland Siegwart 15.06.2017 19
Collaborative Visual-Inertial Navigation teach and repeat https://www.youtube.com/watch?v=pdlqxsorgi4 https://www.youtube.com/watch?v=9pprndikraw Roland Siegwart 15.06.2017 20
Collaborative Transportation master-slave control https://www.youtube.com/watch?v=pdlqxsorgi4 https://www.youtube.com/watch?v=9pprndikraw Roland Siegwart 15.06.2017 21
Complexity of Services Tactile Manipulation Mobile Manipulation Advanced Interaction Autonomous Navigation Actions from simple motion to complex interaction Robotics Roadmap Toys Logistic assistant Industrial services Tour-Guides Transportation logistics Tactile manufacturing Vacuum cleaning Agriculture robots Household universal Search and rescue All-terrain navigation Construction robotics Autonomous car urban Industrial inspection Autonomous car industrial sites Autonomous car freeway 2010 Static Structured, 2D Static Unstructured, 3D Dynamic Structured, 2D Dynamic Unstructured, Dynamic 3D Static Environment - from static 2D grid maps to 3D cognitive maps Semantics dynamic 2030 Roland Siegwart 15.06.2017 22
Switzerland a melting pot for robotics technology Initiatives Spin-offs (*ASL) Industrial Collaborations (ASL) * * * Center for Learning Systems ielab * * * * Roland Siegwart 15.06.2017 23
Tanks to: ASL Team Industrial Partners Funding Agencies Current and former ASL Members (2016) Roland Siegwart 15.06.2017 24