Self-Triggered Model Predictive Control for Signal Temporal Logic Tasks

In this paper, we investigate the reactive control synthesis problem for discrete-time dynamic systems under signal temporal logic (STL) specifications. Our focus is on addressing this problem within the model predictive control (MPC) framework, which involves iteratively solving optimization problems at each time instant. Traditional MPC controllers for STL tasks necessitate sampling the system state at every time instant. To mitigate sensing costs and conserve communication bandwidth during sensor measurement transmission, we introduce a novel concept termed self-triggered MPC for STL tasks. Our proposed approach aims to reduce the overall sampling rate of the system, resulting in considerable energy savings. We provide case studies to demonstrate the effectiveness of the proposed algorithm.