SCS Faculty Candidate

  • Remote Access - Zoom
  • Virtual Presentation - ET

Planning with Humans in the Wild: A Tale of Two Interactions

We are making exciting progress towards deploying robots in the real world, from assembly lines and warehouses to self-driving cars and trucks that run on public roads. However, these robots are still designed with structured scenarios and interactions in mind. If robots are to truly work alongside humans in the wild, they must easily adapt to the diversity of their environments as well as individual human needs. My research aims to build “malleable robots” — robots whose behavior is easy to mold. Such robots can be taught new tasks via natural human interactions and can seamlessly adapt to new environments to execute such tasks.

In this talk, I will explore the two main interactions the robot must engage in: with the environment and with the human. I will show how both of these interactions can be best viewed through the lens of a unified algorithmic framework, where a robot efficiently infers the optimal value function from a bounded set of interactions. This framework ties together insights from both motion planning and imitation learning. I will ground this discussion in concrete examples from my work on autonomous helicopters, robotic manipulation, and self-driving.

Sanjiban Choudhury is a research scientist at Aurora where he works with the best to solve self-driving at scale. He focuses on theory and algorithms at the intersection of machine learning and motion planning. Much of his research has been deployed on real-world robots — full-scale helicopters, self-driving cars, and mobile manipulators. He holds a Ph.D. from The Robotics Institute at Carnegie Mellon University, where he was advised by Sebastian Scherer. His thesis showed how robots can learn from prior experience to speed up online planning. He was a Postdoctoral fellow at the University of Washington, CSE where he worked with Sidd Srinivasa. His research has received best paper awards at ICAPS 2019, finalist for IJRR 2018, and AHS 2014, and winner of the 2018 Howard Hughes award. He is a Siebel Scholar, class of 2013.

Faculty Host: Maxim Likhachev

Robotics Institute

Zoom Participation. See announcement.

For More Information, Please Contact: