Abstract: Against the defect of using traditional ant colony algorithm to solve the problem of indoor evacuation, such as slow convergence, and local optimum, the dynamic parameters of fire field are introduced into the ant colony algorithm, and the path selection strategy, heuristic function and pheromone updating strategy are improved to find the better evacuation path for the whole evacuation group. The improved ant colony algorithm is used to dynamically plan the evacuation path of indoor personnel, considering the real- time congestion degree of the path, avoiding the bottleneck effect of local route optimization of evacuation personnel. The analysis results are compared with the planning results of basic ant colony algorithm. The research results show that the optimization algorithm shortens the evacuation time and planning path, and improves the evacuation efficiency and search speed.

Key words: fire protection, indoor evacuation, ant colony algorithm, congestion, path planning