Adaptive learning controller design for uncertain nonlinear active suspension system with actuator input delay and time-varying state constraints

This paper investigates the adaptive tracking control problem for active suspension system (ASS) with time-varying constraints via the neural network (NN) approach. The nonlinear terms in the spring and damper coefficients, input time delay of the actuator and the suspension performances are also considered simultaneously. The NNs are used to approximate the unknown function caused by the uncertain car-body mass. It is known to all that ensuring the vertical displacement and speed states in the constraint regions can guarantee the stability and safety of the ASSs. Thus, by introducing the symmetric barrier lyapunov function from the time-varying constraints, an adaptive controller based on the NNs and nonlinear backstepping technique is developed to guarantee the states within constraints. Then the stability of the whole system, including the semi-global ultimately uniformly boundedness of the all signals of the ASSs, is analyzed by the Lyapunov stability analysis. Finally, simulation results are provided to verify the feasibility of the adaptive learning controller.