Tire-Road Peak Adhesion Coefficient Estimation Method Based on Fusion of Vehicle Dynamics and Machine Vision

The tire-road peak adhesion coefficient (TRPAC) describes the tire adhesion limit that a road can provide. The TRPAC is a key parameter for precise vehicle motion control and an important basis for decision-making and planning of intelligent vehicles. Considering the critical and difficult problems in the estimation of TRPAC, such as slow convergence and low accuracy, a TRPAC estimation method based on the fusion of vehicle dynamics and machine vision is proposed in this paper. Based on the observability theory of nonlinear systems, local weak observability of the dynamics-based estimator is analyzed to explain the limitation of a single dynamics-based estimator. The framework of dynamics-image-based fusion estimator is then proposed, including the fusion of data, model and decision levels. A dynamics-based fusion estimator is designed by considering the coupling relationship of longitudinal and lateral tire forces to adapt the conditions of complex excitations. Start-and-stop strategy for the dynamics-based fusion estimator is designed by setting excitation thresholds for different types of road surfaces, which are identified using vision information. Parameter self-tuning for the dynamics-based fusion estimator based on the image-based estimator is proposed to improve convergence speed and reduce oscillation. The results of the simulation and vehicle test show that the road estimation error of the proposed method is within 0.03 and the convergence time is within 0.5 s. Compared with other existing estimators, the fusion estimator achieved better accuracy, sensitivity and stability, particularly when complex excitations were present.