Reference-Free Formula Drift with Reinforcement Learning: From Driving Data to Tire Energy-Inspired, Real-World Policies

TL;DR

This paper presents a reinforcement learning approach for autonomous car drifting that uses tire energy concepts, enabling real-world deployment without prior explicit trajectory optimization.

Contribution

It introduces a novel RL-based drifting method trained in simulation with a neural stochastic differential equation vehicle model, achieving zero-shot transfer to real cars.

Findings

Achieved drift tracking error as low as 10 cm

Successfully pushed vehicles to sideslip angles of up to 63°

Demonstrated real-world drifting on Toyota and Lexus vehicles

Abstract

The skill to drift a car--i.e., operate in a state of controlled oversteer like professional drivers--could give future autonomous cars maximum flexibility when they need to retain control in adverse conditions or avoid collisions. We investigate real-time drifting strategies that put the car where needed while bypassing expensive trajectory optimization. To this end, we design a reinforcement learning agent that builds on the concept of tire energy absorption to autonomously drift through changing and complex waypoint configurations while safely staying within track bounds. We achieve zero-shot deployment on the car by training the agent in a simulation environment built on top of a neural stochastic differential equation vehicle model learned from pre-collected driving data. Experiments on a Toyota GR Supra and Lexus LC 500 show that the agent is capable of drifting smoothly through varying waypoint configurations with tracking error as low as 10 cm while stably pushing the vehicles to sideslip angles of up to 63{\deg}.