Learning Mixed Strategies in Trajectory Games

L. Peters, David Fridovich-Keil, L. Ferranti, Cyrill Stachniss, Javier Alonso Mora, Forrest Laine

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

20 Downloads (Pure)

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.”
Original languageEnglish
Title of host publicationProceedings Robotics: Science and System XVIII
EditorsKris Hauser, Dylan Shell, Shoudong Huang
PublisherRobotics Science and Systems (RSS)
Number of pages12
ISBN (Electronic)978-0-9923747-8-5
Publication statusPublished - 2022
EventRobotics: Science and Systems 2022 - New York, United States
Duration: 27 Jun 20221 Jul 2022

Conference

ConferenceRobotics: Science and Systems 2022
Country/TerritoryUnited States
CityNew York
Period27/06/221/07/22

Fingerprint

Dive into the research topics of 'Learning Mixed Strategies in Trajectory Games'. Together they form a unique fingerprint.

Cite this