Research on Formation Reconfiguration of UAVs Based on RRT Algorithm

基于快速扩展随机树算法的多无人机编队重构方法研究

¹ College of Aviation and Foundation, Naval Aviation University, Yantai 264001, China
² College of Aeronautics and Astronautics, National University of Defense Technology, Changsha 410073, China

Received: 8 May 2018

Abstract

To adapt the complexity and flexibility of battlefield environment, a method of formation reconfiguration based on Rapidly-exploring Random Tree (RRT) algorithm is proposed, which shows the advantage of formation of Unmanned Aerial Vehicles (UAVs). Firstly, the kinematic model for UAVs is built, and the feasibility of combination of traditional RRT algorithm and formation reconfiguration of UAVs is analyzed. Secondly, the strategies of trajectory correction comprising node removal and transition trajectory are adopted. Then the dynamic and collision avoidance constraints are discussed respectively, which are essential for exploring the process of RRT algorithm as well as adjusting the trajectory of UAVs. Finally, the simulation and flight experiment are carried out to verify the effectiveness of the proposed method. The results show that the reconfiguration method is able to achieve the formation reconfiguration rapidly and safely. Moreove, the planed trajectory can satisfy the tracing requirement, which is of significance for flight of UAVs in the battlefield environment.

摘要

为适应瞬息万变的战场环境，发挥多无人机不同队形下的作战优势，以快速扩展随机树（RRT）算法为基础，提出一种多无人机编队重构的方法。首先建立多无人机编队的运动模型，分析传统RRT算法与编队重构方法结合的可行性，并采用多余节点去除和构造过渡航迹等策略对航迹进行修正。之后，重点分析重构过程中的动力学及防碰撞等约束，为随机树的扩展和无人机的航迹变换提供依据。最后通过对比仿真和飞行试验，验证所提重构方法的安全性和可行性。结果表明，该重构方法能在复杂环境下快速实现编队重构，同时所规划的航迹利于无人机进行跟踪，可满足实际战场的飞行需求。