摘要

In this paper, a fast human-in-the-loop path planning strategy in cluttered environments based on cloud model is proposed, and it is implemented in a human-machine cooperative Unmanned Aerial Vehicle (UAV) path planning system. Firstly, a dynamic guidance A* (DGA*) search algorithm is proposed to allow human's participation in machine searching loop. Secondly, online uncertainty reasoning based on cloud model is introduced to allow human's fuzzy decision about path direction and trending, then human's perception, expertise, and preferences are incorporated into the DGA* optimality process. Therefore, this effective cooperative decision support can provide a robust solution exploration space, overcoming some shortages of original A* algorithm, such as slow search speed, easily falling into local dead-ends, and so on. Experimental results demonstrate that the proposed method is much more efficient than original A* planner, and generates good solutions that match mission considerations and personal preferences.