intelligent identification of vehicle dynamics based on local model network

In this paper, i propose an intelligent approach for the dynamic identification of vehicles. the developed approach is based upon data-driven identification and uses a high performance local model network (lmn) for estimation of the vehicle longitudinal velocity, lateral acceleration, and yaw rate. the proposed lmn requires no pre-defined standard vehicle model, and uses measurement data to identify vehicle dynamics. lmn is trained by the hierarchical binary tree learning algorithm, which results in a network with maximum generalizability and the best linear or non-linear structure. the proposed approach is applied to a measurement dataset obtained from a volvo v70 vehicle in order to estimate its longitudinal velocity, lateral acceleration, and yaw rate. the identification results reveal that lmn can identify accurately the vehicle dynamics. furthermore, comparison of the lmn results and a multi-layer perceptron neural network demonstrates the far better performance of the proposed approach.