Abstract: In forest fires, rescue robots need to work in a three?dimensional environment, and studying three?dimensional path planning is of great significance. An improved pheromone updating algorithm and a heuristic function with safety value were proposed to solve the common problem of local optimization and long search time in 3D path planning based on ant colony algorithm. To better avoid obstacles and areas with concentrated fire, choose a search mode that combines plane and field of view. According to the test results of the traditional ant colony algorithm and the improved ant colony algorithm, the latter has about 30 iterations less than the former, and the fitness value has decreased by 2.4×10^3, improving the operating efficiency of the system. And the latter’s path selection is more reasonable, proving the superiority of the improved algorithm.

Key words:  rescue robot, ant colony, heuristic function, safety value