Abstract: A novel hybrid path planning algorithm is proposed to solve the problems of long simulation time and low efficiency of evacuation of people in buildings, and the inability to change the evacuation direction according to the fire situation. First, a reward and punishment mechanism is added to the A* algorithm and the heuristic function is improved. Secondly, the problem that the artificial potential field algorithm is easy to fall into local minima is solved by improving the repulsion function. Finally, the two algorithms are combined to form a hybrid path planning algorithm. The feasibility of the hybrid path planning algorithm is verified by modeling and analyzing an educational building with BIM and PyroSim software. The results show that the proposed algorithm can not only find the optimal evacuation path quickly, but also avoid the evacuation path to pass through the dangerous area. Compared with a single algorithm, the hybrid path planning algorithm has fewer traversing nodes, better path smoothness and path planning effect, and higher path planning efficiency.

Key words: path planning, hybrid path planning algorithm, A* algorithm, artificial potential field algorithm