RRTと速度障害物法を組み合わせた無人航空機システム交通管理のための運動計画
RRT and Velocity Obstacles-based motion planning for Unmanned Aircraft Systems Traffic Management (UTM)
RRTでオフライン経路生成、速度障害物法でオンライン衝突回避を行うUTMアルゴリズムを提案し、APFより短い経路長と高い衝突回避性能を示した。
著者: Himanshu, Jinraj V Pushpangathan, Harikumar Kandath
分類: cs.RO
原文アブストラクト
In this paper, an algorithm for Unmanned Aircraft Systems Traffic Management (UTM) for a finite number of unmanned aerial vehicles (UAVs) is proposed. This algorithm is developed by combining the Rapidly-Exploring Random Trees (RRT) and Velocity Obstacle (VO) algorithms and is referred to as the RRT-VO UTM algorithm. Here, the RRT algorithm works offline to generate obstacle-free waypoints in a given environment with known static obstacles. The VO algorithm, on the other hand, operates online to avoid collisions with other UAVS and known static obstacles. The boundary of the static obstacles are approximated by small circles to facilitate the formulation of VO algorithm. The proposed algorithm's performance is evaluated using numerical simulation and then compared to the well-known artificial potential field (APF) algorithm for collision avoidance. The advantages of the proposed method are clearly shown in terms of lower path length and collision avoidance capabilities for a challenging scenario.
