Safe UAV Navigation in Cluttered Environments using MPC-CBFs and Cutting-plane Method

Ensuring the safe navigation of unmanned aerial vehicles (UAV) in cluttered environments presents significant challenges, particularly due to the computational complexity of multiple safety constraints. To address these issues, we propose a unified algorithmic framework that integrates model predictive control (MPC), control barrier functions (CBFs), and a cutting-plane (CP) strategy - termed the MPC-CBFs-CP framework. This framework employs cutting planes to identify and isolate the most critical obstacles, decomposing the obstacle avoidance space and reducing computational burden. Simultaneously, CBFs are embedded within the MPC scheme to enforce safety constraints associated with obstacle avoidance. Simulation results validate the proposed framework's ability to enhance feasibility and computational efficiency while ensuring safe UAV operation in complex environments.