A Gaussian Process Model for Opponent Prediction in Autonomous Racing

TL;DR

This paper introduces a Gaussian process model for predicting opponent behavior in autonomous racing, enhancing safety and performance in overtaking maneuvers through real-time prediction and control in simulation and hardware experiments.

Contribution

It develops a GP-based prediction framework integrated with MPC for autonomous racing, demonstrating improved safety and success rates over existing methods in simulation and real-world tests.

Findings

GP predictor achieves similar win rates with increased safety in simulations

The framework improves overtaking success in hardware experiments

Real-time implementation on a 1/10th scale racecar at 2.8 m/s speeds

Abstract

In head-to-head racing, an accurate model of interactive behavior of the opposing target vehicle (TV) is required to perform tightly constrained, but highly rewarding maneuvers such as overtaking. However, such information is not typically made available in competitive scenarios, we therefore propose to construct a prediction and uncertainty model given data of the TV from previous races. In particular, a one-step Gaussian process (GP) model is trained on closed-loop interaction data to learn the behavior of a TV driven by an unknown policy. Predictions of the nominal trajectory and associated uncertainty are rolled out via a sampling-based approach and are used in a model predictive control (MPC) policy for the ego vehicle in order to intelligently trade-off between safety and performance when attempting overtaking maneuvers against a TV. We demonstrate the GP-based predictor in closed loop with the MPC policy in simulation races and compare its performance against several predictors from literature. In a Monte Carlo study, we observe that the GP-based predictor achieves similar win rates while maintaining safety in up to 3x more races. We finally demonstrate the prediction and control framework in real-time in a experimental study on a 1/10th scale racecar platform operating at speeds of around 2.8 m/s, and show a significant level of improvement when using the GP-based predictor over a baseline MPC predictor. Videos of the hardware experiments can be found at https://youtu.be/KMSs4ofDfIs.