基于改进RRT算法的预警机实时航迹规划  

An Modified RRT-Based Real-Time Route Planning Algorithm for Early Warning Airplane

作　　者：张煜[1] 任保安[1] 陈璟[1] 

机构地区：[1]国防科学技术大学机电工程与自动化学院,湖南长沙410073

出　　处：《计算机仿真》2016年第9期106-112,共7页Computer Simulation

摘　　要：在预警机航迹规划优化问题的研究中,预警机在空中进行科学、精确的实时航迹规划,可提高其实战效能。由于预警系统带有复杂的飞行器动力学约束和路径约束,求解困难,且计算时效性要求高。针对预警机的作战任务要求及载机平台的技战术特性,对经典的RRT算法进行了适应性扩展,提出了一种新的实时航迹规划方法。算法中改进了节点采样方式和节点扩展方式,并采用基于Dubins路径的运动规划方法生成树节点之间的局部最优航迹,使得规划算法不仅能够充分的考虑载机平台的动力学特性,而且规划空间维度远低于常规的搜索空间,算法的时效性得到较大幅度的提升。为了适应动态、不确定的运行环境,算法还可借鉴滚动时域控制(RHC)的思想,设计了在线滚动规划求解框架。仿真结果表明,提出的算法非常符合于复杂、动态环境下的预警机实时航迹规划优化的要求。The key technologies of early warning airplane （EWA） is the autonomous, on-board, real-time route planning, the rapid solution of which is a challenging task. A computationally efficient real-time route planning algorithm based on an improved rapidly-exploring random tree （RRT） algorithm was presented in this paper. To improve the deficiencies of standard RRT algorithm, a simple sampling bias strategy and a new node connection strategy were presented. Then, a motion planner based on Dubins path was used to generate a feasible local optimum trajectory which connects the sample to the selected node in the tree. The use of Dubins path has the advantage of capturing the vehicle＇ s flight limitation with a simple evaluation of an analytical solution, and this planning algorithm samples the reference command rather than the vehicle states or inputs, which typically has a lower dimension, therefore the algo- rithm has high computational efficiency. In order to improve robustness and adaptivity in a dynamic and uncertain environment, a receding horizon planning strategy was introduced as well. Numerical simulations demonstrate the efficiency of this algorithm, and the results also show that the presented technique is well suited for real-time implementation in dynamic and uncertain environment.

关 键 词：实时航迹规划 快速扩展随机树 滚动时域规划 预警机 

分 类 号：V249.1[航空宇航科学与技术—飞行器设计] V271.4