Showing 1–1 of 1 results for author: Blanco, Y

Approximate Robust Control of Uncertain Dynamical Systems

Authors: Edouard Leurent, Yann Blanco, Denis Efimov, Odalric-Ambrym Maillard

Abstract: This work studies the design of safe control policies for large-scale non-linear systems operating in uncertain environments. In such a case, the robust control framework is a principled approach to safety that aims to maximize the worst-case performance of a system. However, the resulting optimization problem is generally intractable for non-linear systems with continuous states. To overcome this… ▽ More This work studies the design of safe control policies for large-scale non-linear systems operating in uncertain environments. In such a case, the robust control framework is a principled approach to safety that aims to maximize the worst-case performance of a system. However, the resulting optimization problem is generally intractable for non-linear systems with continuous states. To overcome this issue, we introduce two tractable methods that are based either on sampling or on a conservative approximation of the robust objective. The proposed approaches are applied to the problem of autonomous driving. △ Less

Submitted 1 March, 2019; originally announced March 2019.

Journal ref: 32nd Conference on Neural Information Processing Systems (NeurIPS 2018) Workshop, Dec 2018, Montr{é}al, Canada