Autonomous Racing using Learning Model Predictive Control

TL;DR

This paper introduces a learning Model Predictive Control approach for autonomous racing that leverages past lap data and vehicle system identification to optimize lap times while ensuring safety, validated through high-fidelity simulations.

Contribution

It presents a novel learning MPC method combined with system identification for autonomous racing, improving lap time performance with safety guarantees.

Findings

Effective lap time reduction demonstrated in simulations

Successful integration of data-driven control and system identification

Validated approach with high-fidelity CarSim simulations

Abstract

A novel learning Model Predictive Control technique is applied to the autonomous racing problem. The goal of the controller is to minimize the time to complete a lap. The proposed control strategy uses the data from previous laps to improve its performance while satisfying safety requirements. Moreover, a system identification technique is proposed to estimate the vehicle dynamics. Simulation results with the high fidelity simulator software CarSim show the effectiveness of the proposed control scheme.