Biasing the Driving Style of an Artificial Race Driver for Online Time-Optimal Maneuver Planning

2025-04-17

AI Summary

This work addresses real-time trajectory planning for autonomous racing, specifically modeling the human driver’s early/late apex selection mechanism and enabling online adaptation of driving style. Method: We propose a novel NMPC terminal cost function leveraging trajectories from prior laps, enabling real-time, continuous, and adaptive biasing toward early or late apex strategies for the first time. The formulation jointly optimizes time-to-completion and exit velocity, integrated within a receding-horizon re-planning framework that extends the prediction horizon while maintaining low computational latency. Contribution/Results: Simulation results demonstrate that the proposed method achieves lap times approaching those of offline global optima—significantly outperforming conventional time-optimal MPC. This validates both the efficacy and practicality of online driving-style adaptation in high-performance autonomous racing.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

In this work, we present a novel approach to bias the driving style of an artificial race driver (ARD) for online time-optimal trajectory planning. Our method leverages a nonlinear model predictive control (MPC) framework that combines time minimization with exit speed maximization at the end of the planning horizon. We introduce a new MPC terminal cost formulation based on the trajectory planned in the previous MPC step, enabling ARD to adapt its driving style from early to late apex maneuvers in real-time. Our approach is computationally efficient, allowing for low replan times and long planning horizons. We validate our method through simulations, comparing the results against offline minimum-lap-time (MLT) optimal control and online minimum-time MPC solutions. The results demonstrate that our new terminal cost enables ARD to bias its driving style, and achieve online lap times close to the MLT solution and faster than the minimum-time MPC solution. Our approach paves the way for a better understanding of the reasons behind human drivers' choice of early or late apex maneuvers.

Problem

• Optimizing artificial race driver's style for time-efficient maneuvers
• Adapting driving style between early and late apex in real-time
• Achieving lap times close to offline optimal solutions

Innovation

• Uses nonlinear MPC for time-optimal trajectory planning
• Introduces MPC terminal cost for driving style adaptation
• Achieves real-time early to late apex maneuver switching