Adaptive robust control for electric linear actuator using modified LuGre model with fast load-based parameter estimation

This paper addresses the position tracking control problem of electric linear actuator in the presence of nonlinearities due to its transmission process. A torque decoupling approach for electric linear actuator is proposed to achieve respective compensation. To improve the speed of time varying parameters estimating, a load-based adaption law is proposed by designing a new projection mapping which has both variable bounds and good robust performance. An adaptive robust controller is designed such that it combines decoupled torque and dead-zone compensation to preserve robustness to both parametric and nonparametric uncertainties. The modified LuGre model is used for friction compensation making a continuous transition from static model. A Lyapunov stability analysis demonstrates that all signals including tracking errors have guaranteed convergent and bounded performance. The co-simulation results show the effectiveness and the achievable control performance of the proposed control strategy with sinusoidal and point-to-point tracking experiments.