Abstract: In case of fire, the existing evacuation route cannot be changed in real time according to the fire situation, which may lead the escape personnel to the fire site, resulting in greater danger. In this paper, a novel improved ant colony algorithm (NIACA) is proposed for fire evacuation path dynamic planning. Firstly, the initial pheromone concentration was improved by A* algorithm, then the equivalent distance improved heuristic function which was affected by fire factors was proposed, then pheromone updating rules were improved to speed up the optimal path search, and finally the path smoothing strategy was used to process the optimal path. The experimental results show that compared with the original ant colony algorithm, the proposed algorithm reduces the blindness in the early stage of the algorithm, has strong dynamic search ability, can help the algorithmavoid falling into local optimization, can quickly and accurately plan the evacuation path in case of fire, and can quickly and safely evacuate the escape personnel to the safe exit away from the fire site.

Key words: fire evacuation path, ant colony algorithm, A*algorithm, equivalent distance