Safe MPC-based Motion Planning of AUVs Using MHE-based Approximate Robust Control Barrier Functions

This paper presents a safety-critical motion plan-ning approach for Autonomous Underwater Vehicles (AUVs). To this end, we formulate a safe Model Predictive Control (MPC) scheme by leveraging the benefits of MPC with a discrete-time Control Barrier Function (CBF). To address the truncation errors associated with the discretized CBF, we then adopt a Moving Horizon Estimation (MHE) approach to approximate these errors. To cope with both the model mismatch in the CBF and unknown external disturbances, we also leverage the MHE scheme to formulate an approximate Robust CBF (RCBF). Finally, to demonstrate the efficacy of the proposed safety-critical robust control scheme, simulation studies are conducted for path planning of AUV s, and the results confirm the superiority of the proposed method compared to few other methods.