Study on Smooth Trajectory Planning for Autonomous Navigation of Quadrotor UAVs in 3D Space for Enhanced Target Recognition
Shibo Chen, Ruinan Fang, Yutao Zhang · 2023
For the flight of autonomous navigation quadrotor UAV, non-smooth paths can lead to rapid rotation or flip of the UAV, which causes problems such as target loss and low detection accuracy. In this paper, combining the kinematic and dynamics models of quadrotor UAV in 3D flight, path planning searches for smooth and smooth trajectories to improve the accuracy of target recognition. The improved Hybrid A* algorithm is applied to the 3D path search of the front-end of the quadrotor UAV. The back end performs nonlinear optimization and smoothing of the trajectory generated by the front-end, and the position, velocity, acceleration and angular velocity of each path point of the path planning are nonlinearly optimized to obtain a smooth, robust and safe trajectory. By improving the robustness of the thrust acting on the UAV, the accuracy of the UAV target identification, detection and reconnaissance is ultimately improved. Compared to the fast planner algorithm, the robustness of the UAV fast flight is improved, the smoothness of the path is increased by a factor of 1.1, the target identification accuracy is increased by a factor of 1.15, and the number of missing targets is reduced by 50%. By improving the heuristic function of Hybrid A* and nonlinear optimization of control points at the back end, the quadrotor UAV is able to fly smoothly and stably and improve the accuracy of target recognition.to identify relevant articles in literature searches, great care should be taken in constructing both.