Agent-based Pathfinding Method for Indoor Fire Emergency Evacuation

doi:10.16182/j.issn1004731x.joss.25-0881

Abstract

Abstract:

To improve emergency evacuation efficiency and reduce casualties in dynamic fire scenarios, a real-time path re-planning method based on agents and dynamic A* algorithm framework is proposed. The behavior and actions of the agent is modeled. A reward function is designed, and an agent-based evacuation framework is constructed. Based on fire simulation data, a dynamic cost network involving parameters such as thermal radiation, smoke visibility, and CO concentration is constructed to achieve spatiotemporally continuous modeling of fire environments. By optimizing the composite cost function through dynamic weight allocation, combined with an improved heuristic function and dynamic search mechanism, local re-planning of evacuation paths and agent-based evacuation pathfinding are achieved. The simulation results show that this method has lower path re-planning response delay and better path safety, demonstrating good adaptability and safety in dynamic hazardous environments.

Key words: dynamic A* algorithm, local re-planning, emergency evacuation, building fire

CLC Number:

TP391.9

Tian Ao, Zhang Jianqin, Wen Zheng, Hu Chaonan, Zhao Hong, Shen bo. Agent-based Pathfinding Method for Indoor Fire Emergency Evacuation[J]. Journal of System Simulation, 2026, 38(2): 532-543.

Figures/Tables 18

Fig. 1

Table 1

Weight values of risk factors in different stages of fire

权重 $ω z$	温度 $ω t$	能见度 $ω v$	CO浓度 $ω C O$
初始期 $ω 1$	0.25	0.55	0.20
蔓延期 $ω 2$	0.50	0.30	0.20
全面期 $ω 3$	0.30	0.20	0.50

Table 1

Table 2

Fig. 2

Fig. 3

Fig. 4

Table 3

Fig. 5

Fig. 6

Fig. 7

Fig. 8

Fig. 9

Fig. 10

Table 4

Fig. 11

Table 5

Table 6

Fig. 12

References 18

[1]	李晓星. 天干物燥, 要防患于未"燃"[N]. 河南日报, 2025-01-09(006).
[2]	倪凌佳, 黄晓霞, 李红旮, 等. 基于协作式深度强化学习的火灾应急疏散仿真研究[J]. 系统仿真学报, 2022, 34(6): 1353-1366.
	Ni Lingjia, Huang Xiaoxia, Li Hongga, et al. Research on Fire Emergency Evacuation Simulation Based on Cooperative Deep Reinforcement Learning[J]. Journal of System Simulation, 2022, 34(6): 1353-1366.
[3]	Wang C, Zhu C, Xiao K, et al. Visualization of Emergency Evacuation Physical Behavior Under Multi-agent Decision-making[J]. Applied Sciences, 2023, 13(9): 5509.
[4]	刘纪坤, 袁雪颖, 王佳茹, 等. 考虑火灾产物影响下的人员疏散路径规划研究[J]. 计算机测量与控制, 2025, 33(4): 155-161.
	Liu Jikun, Yuan Xueying, Wang Jiaru, et al. Research on Personnel Evacuation Path Planning Under the Influence of Fire Products[J]. Computer Measurement & Control, 2025, 33(4): 155-161.
[5]	于萍. 邮轮应急疏散过程建模、模拟与布局优化分析[J]. 中国安全科学学报, 2021, 31(9): 142-149.
	Yu Ping. Modeling and Simulation of Cruise Emergency Evacuation Process for Layout Optimization[J]. China Safety Science Journal, 2021, 31(9): 142-149.
[6]	Zuo Songtao, Mao Zhanli, Fan Chuangang, et al. Dynamic Planning of Crowd Evacuation Path for Metro Station Based on Dynamic Avoid Smoke A-star Algorithm[J]. Tunnelling and Underground Space Technology, 2024, 154: 106145.
[7]	王丽, 李育萌, 刘云, 等. Zigbee耦合A*算法的疏散路径动态规划与指示系统[J]. 中国安全科学学报, 2023, 33(11): 142-149.
	Wang Li, Li Yumeng, Liu Yun, et al. A Dynamic Planning and Indication System for Evacuation Paths Based on Zigbee Coupled a* Algorithm[J]. China Safety Science Journal, 2023, 33(11): 142-149.
[8]	李明海, 兰亚乐, 马骁, 等. 建筑火灾全局动态疏散路径规划研究[J]. 安全与环境学报, 2025, 25(1): 205-215.
	Li Minghai, Lan Yale, Ma Xiao, et al. Research on Global Dynamic Path Planning for Evacuation During Building Fires[J]. Journal of Safety and Environment, 2025, 25(1): 205-215.
[9]	Deng Kunxiang, Zhang Qingyong, Zhang Hang, et al. Optimal Emergency Evacuation Route Planning Model Based on Fire Prediction Data[J]. Mathematics, 2022, 10(17): 3146.
[10]	Ever Enrique Castillo Osorio, Min Song Seo, Hee Yoo Hwan. Consistency of Criteria in Building Interiors Through the Analytic Hierarchy Process for Simulation of Feasible Evacuation Routes[J]. Sensors and Materials, 2020, 32(11): 3835-3851.
[11]	闫俊涛, 秘金钟, 蔚保国, 等. 基于改进A*算法的室内路径规划[J]. 测绘科学, 2023, 48(10): 58-67.
	Yan Juntao, Mi Jinzhong, Yu Baoguo, et al. Indoor Path Planning Based on Improved A* Algorithm[J]. Science of Surveying and Mapping, 2023, 48(10): 58-67.
[12]	蒙盾, 胡卓, 张华军. 基于改进A*算法的多层邮轮疏散系统仿真[J]. 系统仿真学报, 2022, 34(6): 1375-1382.
	Meng Dun, Hu Zhuo, Zhang Huajun. Simulation of Multi-layer Ship Evacuation System Based on Improved A* Algorithm[J]. Journal of System Simulation, 2022, 34(6): 1375-1382.
[13]	林桂娟, 李子涵, 王宇. 基于全局关键点提取的改进A*算法全局路径规划研究[J]. 系统仿真学报, 2025, 37(3): 667-678.
	Lin Guijuan, Li Zihan, Wang Yu. Research on Improved A* Algorithm Path Planning Based on Global Key Point Extraction[J]. Journal of System Simulation, 2025, 37(3): 667-678.
[14]	张海涛, 程荫杭. 基于A*算法的全局路径搜索[J]. 微计算机信息, 2007, 23(17): 238-239, 308.
	Zhang Haitao, Cheng Yinhang. Path Finding Using A*Algorithm[J]. Microcomputer Information, 2007, 23(17): 238-239, 308.
[15]	葛超, 张嘉滨, 王蕾, 等. 一种扩展搜索邻域A*算法的机器人路径规划[J/OL]. 机械设计与制造. (2025-06-04) [2025-09-11]. .
	Ge Chao, Zhang Jiabin, Wang Lei, et al. An Extended Search Neighborhood A* Algorithm for Robot Path Planning[J/OL]. Machinery Design & Manufacture. (2025-06-04) [2025-09-11]. .
[16]	中华人民共和国住房和城乡建设部, 中华人民共和国国家质量监督检验检疫总局. 建筑防烟排烟系统技术标准: [S]. 北京: 中国计划出版社, 2018: 26-27.
	Ministry of Housing and Urban Rural Development of the People's Republic of China, General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China. Technical Standard for Smoke Management Systems in Buildings: [S]. Beijing: China Planning Press, 2018: 26-27.
[17]	张江涛. 基于FDS的办公大楼火灾数值模拟研究[D]. 邯郸: 河北工程大学, 2015.
	Zhang Jiangtao. Numerical Simulation Analysis of Office Building Fire Based on FDS[D]. Handan: Hebei University of Engineering, 2015.
[18]	李引擎. 建筑防火工程[M]. 北京: 化学工业出版社, 2004.

温度范围/℃	能见度范围/m	CO体积分数×10^-6	综合风险场	动态路径代价值	状态
20~30	24~30	0~100	0~0.162	1~2.62	通畅
30~40	18~24	100~250	0.162~0.331	2.26~4.31	正常
40~50	12~18	250~400	0.331~0.509	4.31~6.09	高风险
50~65	8~12	400~550	0.509~0.684	6.09~7.84	极高风险
＞65	—	—	∞	∞	禁行
—	＜8	—	∞	∞	禁行
—	—	＞550	∞	∞	禁行

输入参数	属性
环境温度/℃	20
能见度/m	30
CO浓度	0
风速	0
材料类型	聚氨酯
燃烧表面积/m²	4
模拟时间/s	500

方法	路径效率	距离/m	时间/s	最大通行代价	平均通行代价	平均响应时间/ms	拐点
A*	1.02	60.64	57.66	4.52	2.72	29.60	10
JPS	1.04	61.48	59.49	17.85	3.34	62.39	8
双向A*	1.33	78.70	74.84	37.75	3.14	100.62	11
改进A*	1.02	60.35	56.47	1.36	1.18	8.27	7

方法	路径效率	距离/m	时间/s	最大通行代价	平均通行代价	平均响应时间/ms	拐点
A*	1.11	163.06	154.40	29.58	4.65	102.97	34
JPS	1.04	151.69	181.33	17.62	4.39	276.97	12
双向A*	1.15	168.64	181.36	62.14	21.94	181.36	10
改进A*	1.02	149.38	155.52	11.17	1.39	95.06	8

评价指标	改进A*算法	Pathfinder
ASET/s	137.1	137.1
RSET/s	128.6	130.5
平均疏散距离/m	47.47	55.80
重规划响应时间/ms	70.831	—
出口利用率/%	5~28	1~37