Proxy-Based Super-Twisting Algorithm for MEMS Mirror Control Under Input Saturation and Vibration

Abstract

Microelectromechanical systems (MEMSs) scanning mirrors are widely used in imaging devices such as light detection and ranging (LIDAR) and head-up displays (HUDs). When installed in autonomous vehicles, these MEMS mirrors must accurately track set-points or continuous trajectories despite vibration caused by rugged environments and limited input voltage due to battery management. Super-twisting algorithm (STA), one of the sliding mode control (SMC) strategies with lower chattering effects, can be used to control the MEMS scanning mirror. However, the conventional STA cannot be employed due to requirements for antiwindup and robustness against discontinuous disturbances. This manuscript proposes a novel strategy called proxy-based STA (PSTA), combined with a set-valued terminal SMC, for controlling MEMS scanning mirrors under conditions of discontinuous disturbances and input saturation. The PSTA inherits advantages of the terminal SMC for antiwindup effects and robustness to discontinuous disturbances along with high control accuracy from STA. To reduce chattering effects caused by the terminal SMC, this manuscript further discretizes and realizes PSTA through semi-implicit Euler methods. Both simulation and experimental results demonstrate that this proposed PSTA realization method achieves better performance regarding discontinuous disturbances and input saturation limits.