Energy-based Potential Games for Joint Motion Forecasting and Control

Christopher Diehl, Tobias Klosek, Martin Krueger, Nils Murzyn, Timo Osterburg, Torsten Bertram

Keywords: Trajectory Prediction, Multi-Agent Interaction, Game-Theoretic Motion Planning, Energy-based Model, Optimal Control, Autonomous Vehicles

TL;DR: A connection between differential games, optimal control, and energy-based models and an application for multi-agent motion forecasting and control, combining neural networks with differentiable game-theoretic optimization.

Abstract: This work uses game theory as a mathematical framework to address interaction modeling in multi-agent motion forecasting and control. Despite its interpretability, applying game theory to real-world robotics, like automated driving, faces challenges such as unknown game parameters. To tackle these, we establish a connection between differential games, optimal control, and energy-based models, demonstrating how existing approaches can be unified under our proposed Energy-based Potential Game formulation. Building upon this, we introduce a new end-to-end learning application that combines neural networks for game-parameter inference with a differentiable game-theoretic optimization layer, acting as an inductive bias. The analysis provides empirical evidence that the game-theoretic layer adds interpretability and improves the predictive performance of various neural network backbones using two simulations and two real-world driving datasets.

Student First Author: yes

Instructions: I have read the instructions for authors (https://corl2023.org/instructions-for-authors/)

Video: https://www.youtube.com/watch?v=l6ViD7gvG2o

Code: https://github.com/rst-tu-dortmund/diff_epo_planner

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/energy-based-potential-games-for-joint-motion/code)