关键词: 疏散路径规划, 麻雀搜索算法, 精英反向学习, 余弦算法

Abstract: The environment is complex and changeable when a fire breaks out inside a building, so it is difficult for the traditional evacuation route to make effective route planning according to the actual fire situation. In this paper, an improved sparrow search algorithm is introduced. Firstly, the early warning value parameters of sparrow algorithm are set according to the actual fire location, so as to avoid dangerous areas such as fire sources in the process of path planning. Secondly, the location updating formula of sparrow search algorithm is optimized, and elite reverse learning strategy and sine and cosine optimization algorithm with dynamic weight coefficient are introduced to further improve the problem that sparrow search algorithm is easy to fall into local optimal solution. Finally, the grid method is used to build three types of maps. By comparing the improved sparrow search algorithm with the basic sparrow search algorithm and Grey Wolf algorithm in path planning, it is concluded that the improved sparrow search algorithm has better ability to avoid dangerous areas and path planning in complex fire environment, and is superior to the other two comparison algorithms in path length and number of inflection points, but has a slight deficiency in search time.

Key words: evacuation route planning, sparrow search algorithm, elite reverse learning, cosine algorithm