Adaptive Integral Terminal Sliding Mode Control for Automobile Electronic Throttle via an Uncertainty Observer and Experimental Validation

A robust adaptive integral terminal sliding mode (AITSM) control scheme with an uncertainty observer (UO) is developed for automobile electronic throttle (AET) systems. Distinguished from the ordinary sliding mode (SM)-based AET control systems, by using a novel integral terminal sliding surface (SS) to force the closed-loop system to start on the SS at the very beginning, an excellent error convergence and tracking accuracy as well as robustness of the closed-loop system against uncertainties and nonlinearities including transmission friction, return spring limp-home, and gear backlash can be assured. Also, in order to achieve a stably excellent performance for different tracking commands, the sliding parameters are updated by the designed adaptive laws in Lyapunov sense. Moreover, for reducing the control chattering caused by the large switching gain in conventional SM control, an SM-based UO is introduced to provide with the online estimation for the system lumped uncertainty and conduct the feed-forward compensation in the control design. In addition, model-free robust exact differentiators (MRED) are presented to obtain the accurate angular velocity and acceleration information from the unique angle sensors with measurement noises. Not only the comparative experimental results from a real-time digital signal processor (DSP)-based AET plant but also the quantitive root mean square (RMS) values and the maximum (MAX) values of the sampled tracking errors are demonstrated to validate the superior performance of the proposed control. In particular, in the disturbance rejection test by combining trapezium and sinusoidal reference commands, the proposed control exhibits the smallest RMS error values of 0.6832 deg (37% and 27% of the ones of the nonsingular terminal sliding mode and the proportional-integral-derivative controllers), and the smallest MAX values of 2.9352 deg (67% and 35% of the ones of the comparative controllers, respectively).