Towards Private, Distributed, and Safe Autonomous Systems

Speaker: James Anderson
Affiliation: California Institute of Technology

ABSTRACT: 

Designing controllers to make networked systems autonomous is perhaps the most dominant problem in systems engineering. Distributed control, which is already a challenging task, is only part of the problem. In this talk I will present recently developed theory and algorithms for synthesizing optimal distributed controllers via convex programming.  The theory uses the notion of “locality” in a manner that ensures the synthesis complexity grows gracefully with the number of states in the network model. Recent robust extensions will also be presented.  The remainder of the talk will focus on notions of privacy and safety and how they interact with this notion locality. With respect to privacy, we are concerned with how to build models from data that, for example may contain sensitive information. We present a framework based on aggregation and differential privacy that will encourage users to contribute data while ensuring certain privacy guarantees. From a safety perspective, we will use recent results on invariance to verify that states are always excluded from “unsafe” regions.

BIOGRAPHY: 

James Anderson is a Senior Postdoctoral Scholar in the Department of Computing and Mathematical Sciences at the California institute of Technology.  He obtained a DPhil (PhD) in Engineering Science from the University of Oxford, UK and the BSc and MSc degrees in Systems Engineering from the University of Reading UK.  Prior to Caltech, he held a Junior Research Fellowship in Engineering Science at the University of Oxford and St John’s College, where he was based in the Control Group.  His research interests include distributed control, convex optimization, and privacy preserving computation with applications in autonomy and cyber-physical systems – specifically power grids. He is a lead author of the SOSTOOLS toolbox for polynomial optimization in MATLAB.

For more information, contact Prof. Paulo Tabuada (tabuada@ucla.edu)

Date/Time:
Date(s) - Oct 18, 2018
11:00 am - 12:30 pm

Location:
EE-IV Shannon Room #54-134
420 Westwood Plaza - 5th Flr., Los Angeles CA 90095