一种用于低空飞行的在线航迹重规划方法

研究了飞行器在低空飞行时的在线航迹重规划问题，并提出了一种新方法。通过引入最小威胁曲面的概念生成三维航迹搜索空间，利用一个有限项的多项式函数来逼近最小威胁曲面中的三维航迹在二维水平面内的投影，将航迹规划问题转为在一个一元函数多项式系数空间中的搜索寻优。利用粒子群优化算法完成优化，并借鉴预测控制中滚动时域优化的思想，以在线滚动进行优化的方式完成航迹的在线重规划。实验表明该方法能够规划出合理的航迹并满足实时性要求。

Exploring an Online Method of Vehicle Route Re-Planning for Low Altitude Flight

Up to now, to our best knowledge, papers in the open literature have not yet succeeded in solving the very difficult problem of real-time 3-D vehicle route re-planning to take into account the many uncertainties that actually occur during flight. In this paper, we aim to explore solving this very difficult problem. First of all, the concept of SOMR (Surface of Minimum Risk) is presented to form the search space of 3-D route. Then a polynomial function with finite terms is used to approach the horizontal projection of the 3-D route in the SOMR. So the original planning problem is simplified into that of searching the best series of values in the coefficient space of the polynomial function. This optimization problem is solved with PSO (particle swarm optimization). To realize online re-planning, the idea of receding horizon optimization is introduced to re-plan the route in a way of rolling optimization. This idea and PSO combine to complete the whole planning process. Via the method mentioned above, the generated routes can be re-planned in real-time as new information becomes available, and it has the abilities of terrain following, terrain avoidance and threat avoidance. While no general conclusion can be drawn for the real-time capability of our method, we performed simulations with Matlab 5.3 for a specific flight mission. Simulation results show that: (1) for three conditions, each step of re-planning is estimated to be completed on an average within 2.94 s, 3.31 s and 1.64 s respectively; (2) for re-planning of each step, the allowable time as assumed by us is 10 s.