Tire Slip State Estimation Under Combined Conditions Using Intelligent Tire Technology

In the field of intelligent transportation systems, accurate estimation of tire slip states—specifically slip angle (SA) and slip ratio (SR)—is essential for vehicle safety and performance. However, under combined slip conditions involving both lateral and longitudinal slip, traditional methods struggle due to the nonlinear and highly coupled nature of tire behavior. This study proposes an efficient intelligent tire solution featuring three laterally arranged polyvinylidene fluoride (PVDF) sensors on the tire’s inner liner to capture dynamic strain responses within the contact patch. This layout enables the detection of critical strain variations—tension, compression, and asymmetry—across the tread. From experimental data, 13 physically interpretable features were extracted to model slip states. A Gaussian process regression (GPR) model was developed and compared with artificial neural network (ANN) and support vector machine (SVM) alternatives. Tenfold cross validation confirmed the GPR model’s superior performance, achieving normalized root-mean-square error (NRMSE) values of 2.08% for SA and 1.95% for SR under extreme maneuvers. Critically, this work tackles the underexplored challenge of combined slip estimation by integrating multiregion strain sensing with machine learning. The proposed method achieves promising accuracy ( $\mathrm {NRMSE} \leq 2.08\%$ ), offering a robust framework for intelligent tire systems (ITSs) in complex driving scenarios.