SUNMONTUEWEDTHUFRISAT
Events for April 08, 2019
-
ECE Seminar: A Real-Time Algorithmic Framework for Robust and Risk-Sensitive Planning and Decision-Making
Mon, Apr 08, 2019 @ 11:00 AM - 12:15 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Sumeet Singh, PhD Candidate, Dept of Aeronautics & Astronautics, Stanford University
Talk Title: A Real-Time Algorithmic Framework for Robust and Risk-Sensitive Planning and Decision-Making
Abstract: Integrating autonomous robots into safety-critical settings requires reasoning about uncertainty at all levels of the autonomy stack. In this talk, I will present novel algorithmic tools leveraging Lyapunov-based analysis, convex optimization, and risk measures to address robustness in robotic motion planning and decision-making under uncertainty. In the first part of the talk, by harnessing the theories of incremental stability and contraction, I will describe a unified framework for synthesizing robust trajectory tracking controllers for complex underactuated nonlinear systems with analytical bounded-input-bounded-output disturbance rejection guarantees. These results will be combined with computational tools drawn from semi-infinite convex programming to design real-time motion planning algorithms with certifiable safety guarantees. In addition, I will illustrate how to leverage these tools for sample-efficient model-based reinforcement learning with control-theoretic guarantees. In the second part of the talk, I will describe a framework for lifting notions of robustness from low-level motion planning to higher-level sequential decision-making using the theory of risk measures. Specifically, by leveraging a specific class of risk measures with favorable axiomatic foundations, I will demonstrate how to design decision-making algorithms with tuneable robustness properties. I will then discuss a novel application of this framework to inverse reinforcement learning for humans in safety-critical scenarios. The domains of aerial robotics and autonomous cars will be used throughout the talk as running examples.
Biography: Sumeet Singh is a Ph.D. candidate in the Autonomous Systems Lab in the Aeronautics and Astronautics Department at Stanford University. He received a B.Eng. in Mechanical Engineering and a Diploma of Music (Performance) from University of Melbourne in 2012, and a M.Sc. in Aeronautics and Astronautics from Stanford University in 2015. Prior to joining Stanford, Sumeet worked in the Berkeley Micromechanical Analysis and Design lab at the University of California, Berkeley in 2011 and the Aeromechanics Branch at NASA Ames in 2013. Sumeet's research interests include (1) Robust motion planning for constrained nonlinear systems, (2) Risk-sensitive inference and decision-making with humans in-the-loop, and (3) Design of verifiable learning architectures for safety-critical applications. Sumeet is the recipient of the Stanford Graduate Fellowship (2013-2016), the most prestigious Stanford fellowship awarded to incoming graduate students, and the Qualcomm Innovation Fellowship (2018).
Host: Professor Massoud Pedram, pedram@usc.edu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
-
Fall 2018 Joint CSC@USC/CommNetS-MHI Seminar Series
Mon, Apr 08, 2019 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Jason Marden, University of California, Santa Barbara
Talk Title: If agents could talk, what should they say?
Abstract: The goal in networked control of multiagent systems is to derive desirable collective behavior through the design of local control algorithms. The information available to the individual agents, either through sensing or communication, invariably defines the space of admissible control laws. Hence, informational restrictions impose constraints on the achievable performance guarantees. The first part of this talk will provide one such constraint with regards to the efficiency of the resulting stable solutions for a class of distributed submodular optimization problems. Further, we will also discuss how strategic information exchange can help mitigate these degradations. The second part of this talk will focus on how agents should utilize available information to optimize the efficiency of the emergent collective behavior. In particular, we will discuss a methodology for optimizing the efficiency guarantees (i.e., price of anarchy) in distributed resource allocation problems through the design of local agent objective functions. Lastly, we will highlight some unintended consequences associated with these optimal designed agent objective functions -“ optimizing the performance of the worst-case equilibria (i.e., price of anarchy) often comes at the expense of the best-case equilibria (i.e., price of stability).
Biography: Jason R. Marden is an Associate Professor in the Department of Electrical and Computer, Engineering at the University of California, Santa Barbara. Jason received a BS in Mechanical Engineering in 2001 from UCLA, and a PhD in Mechanical Engineering in 2007, also from UCLA, under the supervision of Jeff S. Shamma, where he was awarded the Outstanding Graduating PhD Student in Mechanical Engineering. After graduating from UCLA, he served as a junior fellow in the Social and Information Sciences Laboratory at the California Institute of Technology until 2010 when he joined the University of Colorado. In 2015, Jason joined the Department of Electrical and Computer Engineering at the University of California, Santa Barbara. Jason is a recipient of the ONR Young Investigator Award (2015), NSF Career Award (2014), the AFOSR Young Investigator Award (2012), the American Automatic Control Council Donald P. Eckman Award (2012), and the SIAM/SGT Best Sicon Paper Award (2015). Furthermore, Jason is also an advisor for the students selected as finalists for the best student paper award at the IEEE Conference on Decision and Control (2011, 2016, 2017). Jason's research interests focus on game theoretic methods for the control of distributed multiagent systems.
Host: Ketan Savla, ksavla@usc.edu
More Info: http://csc.usc.edu/seminars/2019Spring/marden.html
More Information: 190408 Jason Marden CSCUSC Seminar.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 132
Audiences: Everyone Is Invited
Contact: Brienne Moore
Event Link: http://csc.usc.edu/seminars/2019Spring/marden.html
