Regulation-Aware Game-Theoretic Motion Planning for Autonomous Racing

2025-08-27

AI Summary

This work addresses the challenge of motion planning for autonomous racing—balancing safety and competitiveness under real-world racing regulations. We propose RA-GTP, a rule-aware game-theoretic trajectory planning framework. RA-GTP encodes racing rules (e.g., right-of-way, collision liability) as mixed-logical dynamical constraints and embeds them into a generalized Nash equilibrium (GNE) formulation. By integrating model predictive control with an iterative best-response algorithm, RA-GTP enables verifiable, opponent-aware strategic reasoning that explicitly accounts for rule-compliant adversarial behavior. Compared to baseline methods that ignore interaction or regulatory constraints, RA-GTP achieves significantly higher overtaking success rates and more aggressive yet safe maneuvers in simulation—while strictly satisfying all safety and regulatory constraints (100% compliance). To our knowledge, RA-GTP is the first framework to jointly ensure formal rule verifiability, interactive game-theoretic reasoning, and performance optimization in autonomous racing planning.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Multiagent Systems: Multiagent Planning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This paper presents a regulation-aware motion planning framework for autonomous racing scenarios. Each agent solves a Regulation-Compliant Model Predictive Control problem, where racing rules - such as right-of-way and collision avoidance responsibilities - are encoded using Mixed Logical Dynamical constraints. We formalize the interaction between vehicles as a Generalized Nash Equilibrium Problem (GNEP) and approximate its solution using an Iterative Best Response scheme. Building on this, we introduce the Regulation-Aware Game-Theoretic Planner (RA-GTP), in which the attacker reasons over the defender's regulation-constrained behavior. This game-theoretic layer enables the generation of overtaking strategies that are both safe and non-conservative. Simulation results demonstrate that the RA-GTP outperforms baseline methods that assume non-interacting or rule-agnostic opponent models, leading to more effective maneuvers while consistently maintaining compliance with racing regulations.

Problem

• Develops regulation-compliant motion planning for autonomous racing scenarios
• Encodes racing rules using Mixed Logical Dynamical constraints in control
• Solves vehicle interactions as Generalized Nash Equilibrium Problem

Innovation

• Regulation-Compliant Model Predictive Control
• Mixed Logical Dynamical constraints encoding
• Iterative Best Response scheme solution