Robust Transformer-Enhanced Tire Force Estimation: Combining Model-Based Observer and Deep Learning
- 14 January 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 31 (3) , 3582-3593
Abstract
This article proposes a novel hybrid framework for estimating longitudinal tire forces by combining a model-based observer and a Transformer neural network architecture. A key advantage of the proposed approach is its independence from wheel torque measurements, which are commonly used in vehicle force estimation methods. The proposed framework first employs a model-based observer using the Pacejka tire model to generate preliminary estimates of longitudinal tire forces. These estimates are then refined by a neural network featuring an encoder-only Transformer architecture augmented with dilated convolutions, which effectively captures long-term temporal dependencies in sensor data while compensating for model inaccuracies. In addition, our hybrid approach is able to estimate the longitudinal velocity and road friction coefficient, which are crucial for observer’s preliminary force estimate in extreme road and driving conditions. Experimental validation on an electric Equinox vehicle under diverse driving scenarios—including high-slip maneuvers on dry, wet, and icy roads—demonstrates the framework’s superior performance compared to model-based and neural network-only approaches. The proposed method achieves robust estimation accuracy even under sensor faults (e.g., biased acceleration signals) and harsh driving conditions, indicating its potential for enhancing vehicle safety systems, such as traction and stability control.Keywords
All Related Versions
Funding Information
- Natural Sciences and Engineering Research Council of Canada
This publication has 32 references indexed in Scilit:
- A hybrid physics-data driven approach for vehicle dynamics state estimationMechanical Systems and Signal Processing, 2024
- FriC-PM: Machine Learning-based road surface friction coefficient predictive model using intelligent sensor dataConstruction and Building Materials, 2023
- Tire normal force estimation using artificial neural networks and fuzzy classifiers: Experimental validationApplied Soft Computing, 2022
- Hierarchical estimation of vehicle state and tire forces for distributed in-wheel motor drive electric vehicle without previously established tire modelJournal of the Franklin Institute, 2022
- Direct tire slip ratio estimation using intelligent tire system and machine learning algorithmsMechanical Systems and Signal Processing, 2022
- Vehicle state and tyre force estimation: demonstrations and guidelinesVehicle System Dynamics, 2020
- Longitudinal vehicle state estimation using nonlinear and parameter-varying observersMechatronics, 2017
- Corner-based estimation of tire forces and vehicle velocities robust to road conditionsControl Engineering Practice, 2017
- Vehicle state estimation for anti-lock control with nonlinear observerControl Engineering Practice, 2015
- Experimental evaluation of observers for tire–road forces, sideslip angle and wheel cornering stiffnessVehicle System Dynamics, 2008