Learning Mixed Strategies in Trajectory Games

Conference Paper (2022)
Author(s)

L. Peters (TU Delft - Learning & Autonomous Control)

David Fridovich-Keil (The University of Texas at Austin)

L. Ferranti (TU Delft - Learning & Autonomous Control)

Cyrill Stachniss (Universität Bonn)

J. Alonso-Mora (TU Delft - Learning & Autonomous Control)

Forrest Laine (Vanderbilt University)

Research Group
Learning & Autonomous Control
Copyright
© 2022 L. Peters, David Fridovich-Keil, L. Ferranti, Cyrill Stachniss, J. Alonso-Mora, Forrest Laine
DOI related publication
https://doi.org/10.15607/RSS.2022.XVIII.051
More Info
expand_more
Publication Year
2022
Language
English
Copyright
© 2022 L. Peters, David Fridovich-Keil, L. Ferranti, Cyrill Stachniss, J. Alonso-Mora, Forrest Laine
Research Group
Learning & Autonomous Control
ISBN (print)
9780992374785
ISBN (electronic)
978-0-9923747-8-5
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

In multi-agent settings, game theory is a natural framework for describing the strategic interactions of agents whose objectives depend upon one another’s behavior. Trajectory games capture these complex effects by design. In competitive settings, this makes them a more faithful interaction model than traditional “predict then plan” approaches. However, current game-theoretic planning methods have important limitations. In this work, we propose two main contributions. First, we introduce an offline training phase which reduces the online computational burden of solving trajectory games. Second, we formulate a lifted game which allows players to optimize multiple candidate trajectories in unison and thereby construct more competitive “mixed” strategies. We validate our approach on a number of experiments using the pursuit-evasion game “tag.”

Files

P051.pdf
(pdf | 0.746 Mb)
License info not available