Model-free finite-time saturated control for Active vehicle suspension systems with dead zones and external disturbances

Abstract

Active vehicle suspension systems (AVSSs) are important for transportation vehicles to improve ride comfort and maneuverability. However, practical AVSSs normally suffer from uncertain dynamics, unknown external disturbances, input saturations, and dead zones. To address these issues, a novel model-free finite-time saturated control with naturally constrained inputs is proposed for AVSSs to mitigate vibrations and improve ride comfort. Specifically, the hyperbolic function and the bound-based adaptive method are constructed to avoid input saturations. The finite-time convergence can be achieved by designing the nonsingular terminal sliding mode filter. Moreover, the proposed control is a completely model-free approach that does not require any prior knowledge of the exact model information. Therefore, this paper gives the first model-free finite-time saturated control solution for AVSSs that can simultaneously handle input saturations, achieve finite-time convergence, reject external disturbances and uncertain dynamics, maintain model-free structures, and overcome dead zones. The results provide a much improved version of the model-free AVSS control method in which the finite-time stability could be guaranteed with the naturally constrained control input. Various experiments demonstrate the effectiveness and robustness of the proposed algorithm with satisfactory anti-vibration performance and ride comfort (up to 96.5% and 94.8% improvement respectively compared to the passive suspension).