RBFNN based terminal sliding mode adaptive control for electric ground vehicles after tire blowout on expressway

This paper proposes a radial basis function neural network (RBFNN) based terminal sliding mode control scheme for electric ground vehicles subject to tire blowout on expressway in presence of tire nonlinearities, unmodeled dynamics and external disturbances. For enhancing the longitudinal and lateral stability of the vehicle after tire blowout, a saturated velocity planner is firstly constructed for tracking the original motion trajectory, by which the longitudinal velocity and yaw rate saturation constraints can be effectively handled. Afterwards, a terminal sliding mode controller (TSMC) is designed for tracking the planned velocity signals because of its inherent finite time convergence rate and superior steady-state property, by which the adverse dynamic behaviors can be timely suppressed. Further, to strengthen the adaptability and robustness of the control scheme, a RBFNN approximator is developed for identifying the lumped uncertainty, such as tire nonlinearities, unmodeled dynamics and external disturbances, etc., and then compensated into the controller. Lastly, simulations with front-right tire blowout on expressway are performed to validate the effectiveness and efficiency of presented control scheme and methods, and the comprehensive performance of TSMC+RBFNN and TSMC schemes in maintaining original trajectory tracking capacity is evaluated and discussed.