Path planning of UAV based on cultural algorithm in dynamic environments

Due to the randomness of target and threats in complex dynamic environments, it is difficult to plan the path of unmanned aerial vehicle (UAV) in real time. In this paper, we propose path planning of UAV based on cultural algorithm. The belief space consists of situational knowledge and normative knowledge. In the evolutionary process, extract some feature nodes as situational knowledge, and take variation ranges of the nodes as normative knowledge. Once initialized, the path is optimized under the guidance of knowledge in cultural algorithm. When the environment changes or the target moves, there is no need to regenerate the whole path to avoid threat. Only a part of the path, short-path or sub-path, needs to be readjusted. The simulation shows that cultural-algorithm-based path planning can track target and avoid threats rapidly in dynamic environments. Compared with D* algorithm, the proposed method has better real-time performance, lower path planning cost, and shorter length of the optimized path.