AD-MPCC: Adaptive Differentiable Model Predictive Contouring Control for Autonomous Racing

2026-06-30

AI Summary

This work addresses the challenge of simultaneously ensuring safety and achieving high-performance trajectory tracking for autonomous racing vehicles under varying road conditions. We propose the first adaptive framework that integrates online tire-road friction estimation with differentiable Model Predictive Contouring Control (Diff-MPCC). Our approach employs a regularized moving horizon estimator with exponentially decaying weights, combined with the Pacejka “magic formula” to identify friction parameters in real time. A Pacejka-informed neural network, trained via supervised learning, dynamically adjusts the weighting terms in the MPCC cost function. Simulation results demonstrate that, compared to baseline controllers, our method significantly enhances driving safety while effectively reducing lap times across both homogeneous and spatially varying road surfaces.

Abstract

This paper presents Adaptive Differentiable Model Predictive Contouring Control (AD-MPCC), a framework for autonomous racing that integrates differentiable MPCC with online parameter estimation to handle varying road-surface conditions. For online parameter estimation, we leverage a parameterized Pacejka Magic Formula together with a regularized moving-horizon estimation scheme with exponentially decaying weights to capture road interactions and update parameters in real time. Furthermore, we propose a differentiable MPCC (Diff-MPCC) framework that enables optimal adjustment of objective weights based on predefined long-horizon performance costs. To implement Diff-MPCC for online objective weight adaptation, we propose a Pacejka-informed machine learning model that is trained in a supervised manner using data generated by Diff-MPCC to tune the objective weights. Simulation results demonstrate that AD-MPCC reliably ensures safety and achieves faster lap times compared to baseline controllers in both single-surface and multiple-surface scenarios.

Problem

• autonomous racing
• road-surface conditions
• trajectory tracking
• model predictive control
• parameter estimation

Innovation

• Differentiable MPC
• Online Parameter Estimation
• Pacejka Magic Formula
• Adaptive Control
• Autonomous Racing