Print Email Facebook Twitter Learning Mixed Strategies in Trajectory Games Title Learning Mixed Strategies in Trajectory Games Author Peters, L. (TU Delft Learning & Autonomous Control) Fridovich-Keil, David (The University of Texas at Austin) Ferranti, L. (TU Delft Learning & Autonomous Control) Stachniss, Cyrill (Universität Bonn) Alonso-Mora, J. (TU Delft Learning & Autonomous Control) Laine, Forrest (Vanderbilt University) Contributor Hauser, Kris (editor) Shell, Dylan (editor) Huang, Shoudong (editor) Date 2022 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.” To reference this document use: http://resolver.tudelft.nl/uuid:8f70973e-0123-4e7e-865d-a096411c78d2 DOI https://doi.org/10.15607/RSS.2022.XVIII.051 Publisher Robotics Science and Systems (RSS) ISBN 9780992374785 Source Proceedings Robotics: Science and System XVIII Event Robotics: Science and Systems 2022, 2022-06-27 → 2022-07-01, New York, United States Series Robotics: Science and Systems, 2330-765X Bibliographical note Funding Information: This work was supported in part by the National Police of the Netherlands. All content represents the opinion of the authors, which is not necessarily shared or endorsed by their respective employers and/or sponsors. L. Ferranti received support from the Dutch Science Foundation NWOTTW within the Veni project HARMONIA (nr. 18165). Part of collection Institutional Repository Document type conference paper Rights © 2022 L. Peters, David Fridovich-Keil, L. Ferranti, Cyrill Stachniss, J. Alonso-Mora, Forrest Laine Files PDF p051.pdf 763.63 KB Close viewer /islandora/object/uuid:8f70973e-0123-4e7e-865d-a096411c78d2/datastream/OBJ/view