Abstract

This talk addresses the problem of designing motion planning algorithms with rigorous correctness guarantees, with the goal of making planning for autonomous vehicles trusted and certifiable. In the first part of the talk, I will consider the problem of de-randomizing popular sampling-based motion planning algorithms such as the probabilistic roadmap (PRM) algorithm. Randomization, in fact, makes several tasks challenging, including certification and use of offline computation. Leveraging properties of deterministic low-dispersion sequences, I will show that there exist deterministic versions of PRM (and related batch-processing algorithms) that are deterministically asymptotically optimal, enjoy deterministic convergence rates, have improved computational and space complexity properties, and provide superior practical performance.

In the second part of the talk, I will switch to the problem of motion planning under uncertainty. I will present a novel framework whereby motion plans are selected by sampling via Monte Carlo the execution of a reference tracking controller. I will discuss the design of statistical variance-reduction techniques, namely control variates and importance sampling, to make such a sampling procedure amenable to real-time implementation. The advantages of this framework include asymptotic correctness of collision probability estimation and the availability of associated confidence estimates.

I will conclude the talk by discussing applications in the domain of spacecraft autonomous maneuvering.

Speaker Bio

Dr. Marco Pavone is an Assistant Professor of Aeronautics and Astronautics at Stanford University, where he is the Director of the Autonomous Systems Laboratory. Before joining Stanford, he was a Research Technologist within the Robotics Section at the NASA Jet Propulsion Laboratory. He received a Ph.D. degree in Aeronautics and Astronautics from the Massachusetts Institute of Technology in 2010. Dr. Pavone’s areas of expertise lie in the fields of controls and robotics. His main research interests are in the development of methodologies for the analysis, design, and control of autonomous systems, with an emphasis on autonomous aerospace vehicles and large-scale robotic networks. He is a recipient of an NSF CAREER Award, a NASA Early Career Faculty Award, a Hellman Faculty Scholar Award, and was named NASA NIAC Fellow in 2011. He is currently serving as an Associate Editor for the IEEE Control Systems Magazine. His work has been reported in many scientific publications as well as popular press outlets, including ABC, NBC, The Economist, Forbes, and Reuters.