Bayesian Optimization-based Tire Parameter and Uncertainty Estimation for Real-World Data

2025-04-29

AI Summary

This work addresses key challenges in tire parameter identification using real-world autonomous racing data: inaccurate estimation of tire parameters, difficulty in quantifying estimation uncertainty, and lack of assessment of excitation sufficiency. We propose a joint identification method for Pacejka model parameters based on Bayesian optimization and Stochastic Variational Inference (SVI). To our knowledge, this is the first application of SVI to tire modeling, enabling simultaneous estimation of parameter means and their posterior uncertainties. Through systematic sensitivity analysis, we reveal how slip ratio affects the identifiability of individual parameters and establish an excitation sufficiency evaluation criterion. Validation on experimental data from the Abu Dhabi Autonomous Racing League demonstrates that the method significantly improves estimation reliability. Notably, it identifies high posterior uncertainty in curvature and shape parameters under low-slip conditions—directly attributable to insufficient excitation—thereby providing both a data quality assessment framework and practical guidance for parameter selection in vehicle-level tire modeling.

Technology Category

• Reasoning under Uncertainty: Stochastic Optimization
• Machine Learning: Calibration & Uncertainty Quantification
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

This work presents a methodology to estimate tire parameters and their uncertainty using a Bayesian optimization approach. The literature mainly considers the estimation of tire parameters but lacks an evaluation of the parameter identification quality and the required slip ratios for an adequate model fit. Therefore, we examine the use of Stochastical Variational Inference as a methodology to estimate both - the parameters and their uncertainties. We evaluate the method compared to a state-of-the-art Nelder-Mead algorithm for theoretical and real-world application. The theoretical study considers parameter fitting at different slip ratios to evaluate the required excitation for an adequate fitting of each parameter. The results are compared to a sensitivity analysis for a Pacejka Magic Formula tire model. We show the application of the algorithm on real-world data acquired during the Abu Dhabi Autonomous Racing League and highlight the uncertainties in identifying the curvature and shape parameters due to insufficient excitation. The gathered insights can help assess the acquired data's limitations and instead utilize standardized parameters until higher slip ratios are captured. We show that our proposed method can be used to assess the mean values and the uncertainties of tire model parameters in real-world conditions and derive actions for the tire modeling based on our simulative study.

Problem

• Estimating tire parameters and uncertainties using Bayesian optimization
• Evaluating parameter identification quality and required slip ratios
• Assessing real-world data limitations for tire modeling accuracy

Innovation

• Bayesian optimization for tire parameter estimation
• Stochastic Variational Inference for uncertainty assessment
• Real-world data validation in autonomous racing