Stanford Papers and Workshops at RSS 2020

Robotics: Science and Systems (RSS) 2020 is being hosted virtually from July 12th – June 16th. We’re excited to share all the work from SAIL and Stanford Robotics that’s being presented! Below you’ll find links to each paper, as well as the authors’ five-minute presentation of their research. Feel free to reach out to the contact authors directly to learn more about what’s happening at Stanford.

In addition to these papers, SAIL students and faculty are involved in organizing several workshops at RSS. We invite you to attend these workshops, where you can hear from amazing speakers, interact, and ask questions! Workshop attendance is completely virtual: we provide information and links to these workshops at the bottom of this post.

List of Accepted Papers


Shared Autonomy with Learned Latent Actions

Authors: Hong Jun Jeon, Dylan P. Losey, Dorsa Sadigh

Contact: dlosey@stanford.edu

Links: Paper | Experiments

Keywords: human-robot interaction, learning, control


Reinforcement Learning based Control of Imitative Policies for Near-Accident Driving

Authors: Zhangjie Cao*, Erdem Bıyık*, Woodrow Z. Wang, Allan Raventos, Adrien Gaidon, Guy Rosman, Dorsa Sadigh

Contact: caozj@cs.stanford.edu

Links: Paper | Experiments

Keywords: autonomous driving, near-accident scenarios, hierarhical reinforcement learning, hierarchical control, imitation learning


Active Preference-based Gaussian Process Regression for Reward Learning

Authors: Erdem Bıyık*, Nicolas Huynh*, Mykel J. Kochenderfer, Dorsa Sadigh

Contact: ebiyik@stanford.edu

Links: Paper | Experiments

Keywords: active learning, preference-based learning, gaussian processes, reward learning, inverse reinforcement learning


Dynamic Multi-Robot Task Allocation under Uncertainty and Temporal Constraints

Authors: Shushman Choudhury, Jayesh Gupta, Mykel J. Kochenderfer, Dorsa Sadigh, Jeannette Bohg

Contact: shushman@stanford.edu

Links: Paper

Keywords: multi-robot systems, task allocation


GTI: Learning to Generalize across Long-Horizon Tasks from Human Demonstrations

Authors: Ajay Mandlekar*, Danfei Xu*, Roberto Martin-Martin, Silvio Savarese, Li Fei-Fei

Contact: amandlek@stanford.edu

Links: Paper

Keywords: imitation learning, manipulation, robotics


Concept2Robot: Learning Manipulation Concepts from Instructions and Human Demonstrations

Authors: Lin Shao, Toki Migimatsu, Qiang Zhang, Karen Yang, Jeannette Bohg

Contact: lins2@stanford.edu

Links: Paper

Keywords: concept learning; manipulation; learning from demonstration; natural language grounding


ALGAMES: A Fast Solver for Constrained Dynamic Games

Authors: Simon Le Cleac’h, Mac Schwager, Zachary Manchester

Contact: simonlc@stanford.edu

Links: Paper | Blog

Keywords: game theory, optimization, MPC, autonomous driving

List of Organized Workshops


Emergent Behaviors in Human-Robot Systems

Link: Homepage | Zoom (contact ebiyik@stanford.edu for password)

Prerecorded Talks by: Brenna Argall, Anca Dragan, Judith Fan, Jakob Foerster, Robert D. Hawkins, Maja Matarić, Igor Mordatch, Harold Soh, Mac Schwager

Live Events on July 12th (PDT): Panel from 09:30 AM – 10:30 AM | Spotlight talks from 10:30 AM – 11:00 AM

Organizers: Erdem Bıyık, Minae Kwon, Dylan Losey, Noah Goodman, Stefanos Nikolaidis, Dorsa Sadigh


Robotics Retrospectives

Link: Homepage

Prerecorded Talks by: Stefan Schaal. Karen Liu, Sangbae Kim, Andy Zeng, David Hsu, Leslie Kaebling, Nicolas Heess, Angela Schoellig

Live Events on July 12th (PDT): Panels from 09:00 AM – 10:00 AM and 05:00 PM – 06:00 PM | Retrospective sessions from 10:00 AM – 11:00 AM and 04:00 PM – 05:00 PM

Organizers: Franziska Meier, Akshara Rai, Arunkumar Byravan, Jeannette Bohg


Advances & Challenges in Imitation Learning for Robotics

Link: Homepage

Prerecorded Talks by: Anca Dragan, Maya Cakmak, Yuke Zhu, Byron Boots, Abhinav Gupta, Oliver Kroemer

Live Events on July 12th (PDT): Paper discussions from 9:00 AM – 10:30 AM | Virtual coffee break from 10:30 AM – 11:00 AM | Panel from 11:00 AM – 1:30 PM

Organizers: Scott Niekum, Akanksha Saran, Yuchen Cui, Nick Walker, Andreea Bobu, Ajay Mandlekar, Danfei Xu


Action Representations for Learning in Continuous Control

Link: Homepage

Live Talks by: Daniel Braun, Dagmar Sternad, Marc Toussaint, Franziska Meier, Michiel van de Panne

Live Events on July 13th (PDT): Talks from 8:00 AM – 11:50 AM (including spotlights from 9:50 AM – 10:50 AM) | Panel at 11:50 AM

Organizers: Tamim Asfour, Miroslav Bogdanovic, Jeannette Bohg, Animesh Garg, Roberto Martin-Martin, Ludovic Righetti


We look forward to seeing you at RSS!

Read More