Gaussian Process-based Stochastic Model Predictive Control for Overtaking in Autonomous Racing
This addresses the specific challenge of overtaking for autonomous racing systems, representing an incremental advance as it builds on existing lap-time optimization methods.
The paper tackled the problem of planning overtaking maneuvers in autonomous racing by using a Gaussian process to learn the leading vehicle's behavior and a stochastic Model Predictive Control algorithm to plan optimistic trajectories, resulting in successful overtaking in a simple simulation scenario.
A fundamental aspect of racing is overtaking other race cars. Whereas previous research on autonomous racing has majorly focused on lap-time optimization, here, we propose a method to plan overtaking maneuvers in autonomous racing. A Gaussian process is used to learn the behavior of the leading vehicle. Based on the outputs of the Gaussian process, a stochastic Model Predictive Control algorithm plans optimistic trajectories, such that the controlled autonomous race car is able to overtake the leading vehicle. The proposed method is tested in a simple simulation scenario.