Curvature-Constrained UAV Path Planning in Tracking a Moving Air Target
Meng Xu, Lihua Dou, Bin Xin, Yipeng Wang, Hao Fang, Tao Cai · 2018
This paper presents an online motion planning method for an unmanned aerial vehicle (UAV) tracking a moving air target (MAT). The detection range of a UAV is limited to a range of fan-shaped area which is located in the forward viewing angle of the UAV. A mathematical model for the MAT detection and tracking using UAV is built, taking into account the safety distance constraints and the sensor detection range constraints. A Differential Evolution (DE) based method is proposed to solve the above constrained optimization problem. Also, receding horizon control (RHC) is adopted to improve the efficiency of DE. Moreover, in order to ensure the generated paths smooth and flyable, Dubins path planning method is used, in consideration of curvature constrains. The impact of sensor detection error on path planning is considered and analyzed. Furthermore, we employ a widely used optimization method Genetic Algorithm (GA) as a competitor. The calculation and comparison results show that the DE performs better in detecting and tracking the MAT, in the constrast to GA, and it can generate feasible path much more efficiently.