Fine-tuning hybrid dynamics with physics-informed neural networks for vehicle dynamics estimation
Authors:
Open Access
- 25 April 2025
- journal article
- research article
- Published by Springer Nature in International Journal of Intelligent Robotics and Applications
- Vol. 9 (4) , 1594-1610
Abstract
Accurate dynamic modeling is critical for autonomous racing vehicles, especially during high-speed and agile maneuvers where precise motion prediction is essential for safety. Traditional parameter estimation methods face limitations such as reliance on initial guesses, labor-intensive fitting procedures, and complex testing setups. On the other hand, purely data-driven machine learning methods struggle to capture inherent physical constraints and typically require large datasets for optimal performance. To address these challenges, this paper introduces the Fine-Tuning Hybrid Dynamics (FTHD) method, which integrates supervised and unsupervised Physics-Informed Neural Networks, combining physics-based modeling with data-driven techniques. FTHD fine-tunes a pre-trained Deep Dynamics Model (DDM) using a smaller training dataset, delivering superior performance compared to state-of-the-art methods such as the Deep Pacejka Model and outperforming the original DDM. Furthermore, an Extended Kalman Filter (EKF) is embedded within FTHD (EKF-FTHD) to effectively manage noisy real-world data, ensuring accurate denoising while preserving the vehicle’s essential physical characteristics. The proposed FTHD framework is validated through scaled simulations using the BayesRace Physics-based Simulator and full-scale real-world experiments from the Indy Autonomous Challenge. Results demonstrate that the hybrid approach significantly improves parameter estimation accuracy, even with reduced data, and outperforms existing models. EKF-FTHD enhances robustness by denoising real-world data while maintaining physical insights, representing a notable advancement in vehicle dynamics modeling for high-speed autonomous racing.Keywords
Funding Information
- National Science Foundation (CMMI 2146056)
- National Science Foundation (CMMI 2146056)
This publication has 26 references indexed in Scilit:
- Fine-Tuning Hybrid Physics-Informed Neural Networks for Vehicle Dynamics Model EstimationIFAC-PapersOnLine, 2024
- Data-driven deep learning approach for thrust prediction of solid rocket motorsMeasurement, 2023
- Physics Informed Neural Networks: Reducing Data Size Requirements via Hybrid Learning2022
- Indy Autonomous Challenge - Autonomous Race Cars at the Handling Limits2022
- A Dual-Dimer method for training physics-constrained neural networks with minimax architectureNeural Networks, 2021
- Computing the racing line using Bayesian optimization2020
- End-to-End Neural Network for Vehicle Dynamics Modeling2020
- Information theoretic MPC for model-based reinforcement learning2017
- Parameter identification in dynamic systems using the homotopy optimization approachMultibody System Dynamics, 2011
- THE MAGIC FORMULA TYRE MODELVehicle System Dynamics, 1992