Autonomous Drifting with 3 Minutes of Data via Learned Tire Models
This work addresses safety in emergency vehicle maneuvers by enabling efficient autonomous drifting, though it is incremental as it builds on existing control frameworks.
The paper tackled the problem of modeling tire forces near adhesion limits for autonomous drifting by proposing a neural ODE-based tire model, achieving a 4x improvement in tracking performance and enabling high-performance drifting with less than three minutes of data at speeds up to 45mph.
Near the limits of adhesion, the forces generated by a tire are nonlinear and intricately coupled. Efficient and accurate modelling in this region could improve safety, especially in emergency situations where high forces are required. To this end, we propose a novel family of tire force models based on neural ordinary differential equations and a neural-ExpTanh parameterization. These models are designed to satisfy physically insightful assumptions while also having sufficient fidelity to capture higher-order effects directly from vehicle state measurements. They are used as drop-in replacements for an analytical brush tire model in an existing nonlinear model predictive control framework. Experiments with a customized Toyota Supra show that scarce amounts of driving data -- less than three minutes -- is sufficient to achieve high-performance autonomous drifting on various trajectories with speeds up to 45mph. Comparisons with the benchmark model show a $4 \times$ improvement in tracking performance, smoother control inputs, and faster and more consistent computation time.