Motion Planning and Control of Quadrotor Based on MPC in Frenet Frame

The autonomous flight of Unmanned Aerial Vehicles (UAVs) hinges on the interplay between trajectory planning and flight control. This study centers on a specific case: the autonomous flight of a quadcopter at low altitudes. To enhance this autonomous flight process, we address the estimation and prediction of trajectories for dynamic obstacles that may intersect the UAV's low-altitude flight path. This is achieved through the utilization of Extended Kalman Filtering (EKF) and Joint Probabilistic Data Association (JPDA) techniques within the Frenet coordinate system. Subsequently, leveraging both global planning and real-time environmental data, the local motion planner constructs an optimal trajectory for the UAV on a local scale. This ensures that the UAV navigates its immediate surroundings efficiently. To execute this trajectory, the quadrotor UAV undertakes low-altitude flight while adhering to the locally optimized path. This is accomplished through the application of Model Predictive Control (MPC), which enables the UAV to autonomously avoid collisions in real time. Finally, the validity and viability of the proposed approach are substantiated.