Event-triggered sliding mode predictive control for trajectory tracking of space robots with multiple disturbance and noisy position measurements

Space robots can complete the various space missions, which is promising for future space applications. However, the existence of disturbances, model uncertainties, measurement noise, and different constraints would degrade control performance, even leading to self-collision. In response to these challenges, this paper presents an event-triggered sliding mode predictive control approach for trajectory tracking of space robots. The extended Kalman filter enhanced disturbance observer is used to obtain high accuracy estimation results, which can improve control performance. Additionally, the proposed sliding mode predictive control framework is structured in two loops: an outer loop employing event-triggered linear time-varying model predictive control to handle the constraints such as bounded control torque and maximum rotation velocity, and an inner loop utilising discrete fast terminal sliding mode control to further handle the bounded control torque constraint and improve robustness. Numerical simulations demonstrate the effectiveness, feasibility, and robustness of the approach.