Tim Barfoot (University of Toronto) - Towards Reliable and Generalizable Robot Navigation скачать в хорошем качестве

Tim Barfoot (University of Toronto) - Towards Reliable and Generalizable Robot Navigation

Tim Barfoot (University of Toronto) Towards Reliable and Generalizable Robot Navigation Presentation was recorded March 6, 2020, at Vector Institute For the last decade, we have been working on sensor-based navigation of mobile robots for a variety of applications. Our most successful technique, called Visual Teach & Repeat (VT&R), works well in practice because it (i) exploits human experience in route definition, (ii) avoids the need to build a global map of the world, and (iii) plays to the strengths of computer vision by keeping the viewpoints the same between teach and repeat. However, to scale up to real-world operations, we need to be able to generalize across sensors (camera, lidar, radar), robots (driving, flying, walking, swimming), and scenes (lighting, weather, geometry, dynamic obstacles) to a level well beyond our current capabilities. In this talk, I will introduce the VT&R concept, what we’ve been able to accomplish so far, and the current limitations. Moving forward, we are interested in how to build a general template for a navigation solution, then learn the robot/task-specific details from data. I will talk about some of our steps in this direction and some of the challenges we are facing. There will be lots of videos (of robots). Biography: Prof. Timothy Barfoot (University of Toronto Institute for Aerospace Studies – UTIAS) works in the area of autonomy for mobile robots targeting a variety of applications. He is interested in developing methods (localization, mapping, planning, control) to allow robots to operate over long periods of time in large-scale, unstructured, three-dimensional environments, using rich onboard sensing (e.g., cameras and laser rangefinders) and computation. Tim holds a BASc (9T7+PEY, Aerospace Option) from the UofT Engineering Science program and a PhD from UTIAS in robotics. He took up his academic position at UTIAS in May 2007, after spending four years at MDA Robotics (builder of the well-known Canadarm space manipulator), where he developed autonomous vehicle navigation technologies for both planetary rovers and terrestrial applications such as underground mining. Tim was also a Visiting Professor at the University of Oxford in 2013 and recently completed a leave as Director of Autonomous Systems at Apple in California in 2017-9. He is currently the Chair of the UofT Engineering Science Robotics Option, Associate Director of the UofT Robotics Institute, and a Faculty Affiliate of the Vector Institute. He sits on the editorial boards of the International Journal of Robotics Research (IJRR) and the Journal of Field Robotics (JFR) and served as the General Chair of Field and Service Robotics (FSR) 2015, which was held in Toronto. He is the author of a book, State Estimation for Robotics (2017), which is free to download from his webpage (http://asrl.utias.utoronto.ca/~tdb).