Vehicle stability control is crucial for securing vehicle driving safety, while precise stability classification can assist in enhancing the performance of vehicle control. A stable adaptive model predictive controller (MPC) of the four-wheel independent steering (4WIS) vehicle is introduced in this paper. Firstly, a new vehicle dynamic model derived from a long short-term memory (LSTM) network is established, and an attribute dataset representing vehicle stability is procured. Subsequently, we employ the Gaussian mixture model (GMM)-hidden Markov model (HMM) to classify the stability of the 4WIS vehicle. According to the different classification results, the stable adaptive MPC integrated with Bayesian optimisation (BO) is designed to track an optimal trajectory with high accuracy while working under different conditions. Through the simulation tests in various typical driving scenarios, the advantages of the stability classification strategy and the stable adaptive control method proposed in this paper are confirmed.
Pan et al. (Wed,) studied this question.