Adaptive fixed-time dynamic sliding mode control via prescribed performance for automotive electronic throttle valves

This paper addresses the high-performance control problem for automotive electronic throttle valve (ETV) systems under nonlinearities, uncertainties, disturbances, and input saturation. Existing methods suffer from three critical limitations: prescribed performance control only ensures asymptotic convergence with initial-condition-dependent settling time, fixed-time controllers lack integration with engineering performance specifications, and input saturation is treated separately without unified compensation. To overcome these, we propose an adaptive fixed-time dynamic sliding mode control with prescribed performance (AFSMC-PP). The contributions lies in four aspects: (i) an improved fixed-time performance function that guarantees convergence within a fixed time independent of initial conditions, differing from asymptotic PPC. (ii) a synergistic combination of integral-type dynamic sliding surface and fixed-time disturbance observer for chattering suppression and rapid uncertainty compensation. (iii) explicit input saturation compensation via an auxiliary system, resolving the high-gain saturation issue in PPC. and (iv) adaptive laws with dead-zone modification eliminating prior knowledge of uncertainty bounds. Lyapunov-based analysis proves fixed-time stability and prescribed performance. Hardware-in-the-loop experiments show that AFSMC-PP reduces maximum absolute error by up to 46.9% and 72.6%, and RMSE by up to 21.8% and 70.0%, compared to existing adaptive finite-time and adaptive prescribed performance controllers. This work provides a unified, practically viable solution reconciling saturation, fixed-time convergence, and strict transient/steady-state performance for ETV control.