Check out the UofT contributions to the 2020 Robotics Science and Systems conference.

Workshops

Structured Approaches to Robot Learning for Improved Generalization

Organized by Arunkumar Byravan, Markus Wulfmeier, Franziska Meier, Mustafa Mukadam, Nicolas Heess, Angela Schoellig, Dieter Fox

Power On and Go Robots: ‘Out-of-the-Box’ Systems for Real-World Applications

Organized by Jonathan Kelly,  Stephan Weiss, Robuffo Giordana, Valentin Peretroukhin

The new service robot you ordered has just arrived! You take it out of the box, flip the “on” switch and… and… it sits immobile, the animated face on its screen looking confused. Frustrated, after an hour of trying to coax it to do something, you call the company and ask for a refund.

Action Representations for Learning in Continuous Control 

Organized by Tamim Asfour, Miroslav Bogdanovic, Jeannette Bohg, Animesh Garg, Roberto Martín-Martín, Ludovic Righetti

In this workshop we study a new paradigm that bridges robot learning and advanced robot control. We will try to shed light on questions such as: what properties of manipulation tasks indicate the best action space for robot learning? How to best incorporate compliant and force control into learning algorithms? Do the most common action spaces emerge as natural representation to learn tasks?

Workshop on Visual Learning and Reasoning for Robotic Manipulation

Co-organized by Kuan Fang, David Held, Yuke Zhu, Dinesh Jayaraman, Animesh Garg, Lin Sun, Yu Xiang, Greg Dudek

Papers

A Smooth Representation of Belief over SO(3) for Deep Rotation Learning with Uncertainty — Best Student Paper Award!

Co-authored by Valentin Peretroukhin, Matthew Giamou, David M. Rosen, W. Nicholas Greene, Nicholas Roy, Jonathan Kelly

Vision-Based Goal-Conditioned Policies for Underwater Navigation in the Presence of Obstacles

Travis Manderson (McGill University); Juan Camilo Gamboa Higuera (McGill University); Stefan Wapnick (McGill University); Jean-François Tremblay (McGill University); Florian Shkurti (University of Toronto); David Meger (McGill University); Gregory Dudek (McGill University)

Workshop Talks

Robust Autonomy Workshop

One Loss to Rule them All: Towards End-t0-End Autonomy Design Using Variational Inference, by Tim Barfoot

Retrospectives in Robotics Workshop

CTCR Prototype Development: An Obstacle in the Research Community?, co-authored by Reinhard M. Grassman, Sven Lilge, Phung H. U. Le, Jessica Burgner-Kahrs