关键词: 综合建筑, 消防疏散图, 蚁群算法, 路径规划

Abstract:

China is currently in a period of rapid urban development, the comprehensive buildings continue to emerge. At this time, an intelligent fire evacuation map that can automatically adjust the evacuation path based on real-time fire information emerges as the times require. Path planning algorithm is one of the key research directions of intelligent fire evacuation map. Aiming at the problems of intelligent fire evacuation maps that need to be combined with fire field information to dynamically plan paths, this paper proposes an improved bidirectional ant colony algorithm. First, this paper adds bidirectional search strategy to improve the algorithm’s global search capability. Then this paper combines with the A*algorithm to improve the initial pheromone distribution and reduces the blindness of the initial search of the algorithm. In order to improve the convergence speed of the algorithm, the pheromone update strategy is improved. Finally, this paper combines the fire scene information and steering penalty coefficient, improves algorithm evaporation coefficient, heuristic function and transition probability. The risk of the algorithm falling into the local optimum is reduced, improving the algorithm search efficiency and path smoothness, and effectively avoiding the fire affected area. Simulation results show the effectiveness of the algorithm.

Key words: comprehensive building, fire evacuation plans, ant colony algorithm, route plan