Mixed-Integer MPC for Closed-Loop Motion Planning and Flight Control of a Laboratory Helicopter

This work addresses the problem of planning and tracking collision-free trajectories with a laboratory helicopter restrained to three degrees of freedom (DOF) using model predictive control (MPC) with mixed-integer programming (MIP) encoding. We propose a multirate structure that divides the system dynamics into slower and faster modes. The MPC-MIP is then employed to directly control only the slower portion, allowing it to operate in a closed loop in spite of its relatively time-demanding optimizations. Experimental results show that the proposed control structure allows the computation of globally optimal trajectories and controls with the required frequency, resulting in the successful maneuvering of the helicopter toward a target in an obstacle-filled environment.