主管：中华人民共和国应急管理部
主办：应急管理部天津消防研究所
ISSN 1009-0029  CN 12-1311/TU
消防科学与技术

消防科学与技术 ›› 2023, Vol. 42 ›› Issue (10): 1386-1392.

• • 上一篇    下一篇

基于GDACO算法的引导疏散路径规划研究

王馨1, 张林鍹1,2, 董芳芳3   

  1. （1. 新疆大学 电气工程学院，新疆 乌鲁木齐 830017；2. 清华大学 国家计算机集成制造系统工程技术研究中心，北京 100084；3. 新疆维吾尔自治区消防救援总队，新疆 乌鲁木齐 830000）
  • 出版日期:2023-10-15 发布日期:2023-10-15
  • 作者简介:王 馨（1998- ），女，新疆伊犁人，新疆大学电气工程学院硕士研究生，主要从事人员疏散路径规划研究，新疆维吾尔自治区水磨沟区华瑞街777号，830017。

Guided evacuation path planning research based on the GDACO algorithm

Wang Xin1, Zhang Linxuan1,2, Dong Fangfang3   

  1. (1. School of Electrical Engineering, Xinjiang University, Xinjiang Urumqi 830017, China; 2. National Computer Integrated Manufacturing System Engineering Research Center, Tsinghua University, Beijing 100084, China; 3. Xinjiang Uygur Autonomous Region Fire and Rescue Headquarters, Xinjiang Urumqi 830000, China)
  • Online:2023-10-15 Published:2023-10-15

摘要： 针对教学建筑内中人员疏散存在盲目、无序、效率低等问题，提出了双层蚁群路径规划算法（GDACO）引导人员疏散的路径规划。对环境栅格进行处理，增添危险栅格，使初始蚁群信息素呈阶梯式分布；提出双层蚁群算法，引导层蚁群通过全局搜索寻找引导路径，疏散层蚁群基于引导范围实现疏散路径寻优；进行路径优化处理，剔除冗余节点，对转折部分进行平滑处理。试验结果分析表明，所提GDACO算法能够减少路径转折点，规避风险路径，满足教学建筑内的引导疏散需求，规划路径符合实际疏散情况。

关键词: 疏散引导, 路径规划, 信息素分布, 双层蚁群算法, 路径优化, 教学建筑

Abstract: For the problems of blind, disorderly, and inefficient evacuation of personnel in teaching buildings, the Guided Double Layer Ant Colony Optimization Algorithm （GDACO） is proposed for guiding personnel evacuation. Firstly, the environmental grids are processed by adding dangerous grids to establish a staggered distribution of initial ant pheromones. Secondly, the double layer ant colony algorithm is introduced, where the guiding layer ants conduct global searches to find guiding paths while the evacuation layer ants optimize evacuation paths based on the guiding range. Finally, path optimization is performed by removing redundant nodes and smoothing turning points. Experimental results analysis shows that the proposed GDACO algorithm can reduce path turning points, avoid risky paths, meet the evacuation needs within teaching buildings, and plan paths that align with actual evacuation situations.

Key words: evacuation guidance, path planning, pheromone distribution, double layer ant colony algorithm, path optimization, teaching building