面向复杂火灾场景的AI驱动无人化消防救援体系综述

doi:10.16265/j.cnki.issn1003-3033.2026.07.0505

中国安全科学学报 ›› 2026, Vol. 36 ›› Issue (7): 277-286.doi: 10.16265/j.cnki.issn1003-3033.2026.07.0505

• 智能安全 • 上一篇

面向复杂火灾场景的AI驱动无人化消防救援体系综述

王佩¹^,²(), 何昌原³^,^**(), 尹燕博², 王建伟², 黄思聪², 王港¹

¹ 北京师范大学未来技术学院, 广东珠海 519087
² 中国消防救援学院消防指挥系, 北京 102202
³ 中国民航管理干部学院机场管理系, 北京 100102

收稿日期:2026-03-24 修回日期:2026-05-27 出版日期:2026-07-28
通信作者:
^**何昌原(1996—),男,吉林长春人,博士,讲师,主要从事低空安全、无人机应急救援等方面的研究。E-mail:hechangyuan@camic.cn。
作者简介:
王佩 (1991—),男,山西曲沃人,博士研究生,讲师,主要研究方向为消防指挥、应急救援技术。E-mail:wangpeiwf@cfri.edu.cn。
尹燕博,讲师;
王建伟, 讲师;
王港, 副研究员
基金资助:
国家消防救援局科技计划项目(2024XFCX42); 中国消防救援学院院级项目(XFKYB202509); 中国民航管理干部学院青年教师科研启动基金资助(25QN03); 中国博士后科学基金特别资助(2025T180861)

A review of AI-driven unmanned rescue framework for complex fire scenarios

Wang Pei¹^,²(), He Changyuan³^,^**(), Yin Yanbo², Wang Jianwei², Huang Sicong², Wang Gang¹

¹ School of Future Technology, Beijing Normal University, Zhuhai Guangdong 519087, China
² Department of Fire Command, China Fire and Rescue Institute, Beijing 102202, China
³ Department of Airport Management, Civil Aviation Management Institute of China, Beijing 100102, China

Received:2026-03-24 Revised:2026-05-27 Published:2026-07-28

摘要/Abstract

摘要：

为提高消防安全,针对复杂火场面临的感知滞后、近身作业风险高与异构装备协同弱等救援瓶颈,以及通用人工智能(AI)向消防场景迁移时存在的环境适配差、动态调度弱与人机协同机制不清等阻碍,综合采用文献计量分析、综述等方法,剖析2016—2025年通用AI无人化技术领域与消防无人化领域相关研究的演进差异。结果表明:通用AI无人化技术领域相关年发文总量远超消防细分领域,两者研究成熟度存在明显的技术剪刀差;复杂火场救援的关键约束已由单一平台性能不足,转向多目标优化、断续通信条件下的协同控制以及可信决策治理。基于系统工程视角,提出面向复杂火场的“五层八步”无人化体系架构,从终端执行、边缘智能、协同控制、通信网络和指挥孪生5个层级,构建感知-理解-决策-传输-执行-评估-学习-优化的业务闭环,该架构在局部安全保护、任务连续性与链路恢复后重调度方面显著优于传统云-端架构。在此基础上,凝练出多模态抗干扰感知与受限算力下的边缘推理等4项技术方向,以及资源配置与动态调度耦合等3类前沿科学问题。

关键词: 复杂火灾场景, 消防救援, 人工智能(AI)驱动, 无人化体系, 数字孪生

Abstract:

To enhance fire safety, the rescue bottlenecks in complex fire scenes were addressed, including perception lag, high risks of close-range operations, and weak coordination among heterogeneous equipment. The obstacles to migrating general AI into firefighting scenarios were also considered, such as poor environmental adaptability, deficient dynamic scheduling, and ambiguous human-machine collaboration. Bibliometric analysis and systematic review were jointly employed. The evolutionary divergence between the general field of AI-enabled unmanned technologies and research related to unmanned firefighting from 2016 to 2025 was thereby characterized. The results showed that the annual publication volume in the general AI-enabled unmanned technology domain far exceeded that in the firefighting subdomain. A significant gap in technological maturity was observed between the two. The critical constraints of complex fire rescue had shifted from single-platform performance limitations toward multi-objective optimization, collaborative control under intermittent communication, and trustworthy decision governance. From a systems engineering perspective, a five-layer eight-step unmanned architecture for complex fire scenes was proposed. Five layers were defined, namely terminal execution, edge intelligence, collaborative control, communication network, and command digital twin. An operational closed-loop process was constructed across these layers, covering perception, understanding, decision, transmission, execution, evaluation, learning, and optimization. The proposed architecture markedly outperformed the conventional cloud-terminal paradigm in local safety protection, task continuity, and post-recovery rescheduling. On this basis, four technical directions, including multimodal anti-interference perception and edge inference under constrained computing resources, together with three categories of emerging research questions, such as the coupling of resource allocation and dynamic scheduling, were further distilled.

Key words: complex fire scenarios, fire rescue, artificial intelligence (AI)-driven, unmanned rescue framework, digital twin

中图分类号:

X928.7火灾与爆炸事故

王佩, 何昌原, 尹燕博, 王建伟, 黄思聪, 王港. 面向复杂火灾场景的AI驱动无人化消防救援体系综述[J]. 中国安全科学学报, 2026, 36(7): 277-286.

Wang Pei, He Changyuan, Yin Yanbo, Wang Jianwei, Huang Sicong, Wang Gang. A review of AI-driven unmanned rescue framework for complex fire scenarios[J]. China Safety Science Journal, 2026, 36(7): 277-286.

图/表 8

图1

图2

图3

表1

表2

图4

图5

表3

参考文献 26

[1]	张丽平, 李春红, 李松. 针对室内火灾应急响应的空间关键字查询方法研究[J]. 中国安全生产科学技术, 2024, 20(11): 118-123.
	Zhang Liping, Li Chunhong, Li Song. Research on spatial keyword query methods for indoor fire emergency response[J]. Journal of Safety Science and Technology, 2024, 20(11): 118-123.
[2]	王起全, 刘志刚, 杨鑫刚, 等. 化工园区重大危险源风险情景构建分析[J]. 中国安全科学学报, 2020, 30(8): 63-69. doi: 10.16265/j.cnki.issn1003-3033.2020.08.010
	Wang Qiquan, Liu Zhigang, Yang Xingang, et al. Scenario construction analysis of major hazard sources in chemical industry parks[J]. China Safety Science Journal, 2020, 30(8): 63-69. doi: 10.16265/j.cnki.issn1003-3033.2020.08.010
[3]	王秉, 史志勇. 数智应急文化:数智应急时代应急文化的新形态[J]. 复旦公共行政评论, 2025(1): 74-95.
	Wang Bing, Shi Zhiyong. Digital-intelligent emergency culture: a new form of emergency culture in the era of digital-intelligent emergency[J]. Fudan Public Administration Review, 2025(1): 74-95.
[4]	陈涛, 黄丽达, 陈杨, 等. 公共安全领域工程科技未来20年发展战略研究[J]. 中国工程科学, 2024, 26(5): 65-79. doi: 10.15302/J-SSCAE-2024.05.008
	Chen Tao, Huang Lida, Chen Yang, et al. Development strategy of public safety engineering science and technology in the next 20 years[J]. Strategic Study of CAE, 2024, 26(5): 65-79. doi: 10.15302/J-SSCAE-2024.05.008
[5]	王秉, 史志勇. 数智赋能情报驱动的城市低空安全态势感知塑造机制[J]. 城市发展研究, 2025, 32(6): 9-17,35.
	Wang Bing, Shi Zhiyong. Research on the mechanism of digital intelligence-enabled and intelligence-driven urban low-altitude safety & security situation awareness and shaping[J]. Urban Development Studies, 2025, 32(6): 9-17,35.
[6]	王勇, 胡斌, 康渴楠, 等. 智能传感器在消防装备中的应用现状分析[J]. 中国安全科学学报, 2022, 32(增刊1): 184-188.
	Wang Yong, Hu Bin, Kang Kenan, et al. Analysis of application status of intelligent sensors in firefighting equipment[J]. China Safety Science Journal, 2022, 32(S1): 184-188. doi: 10.16265/j.cnki.issn1003-3033.2022.S1.0657
[7]	段汝甫, 佟润泽. 基于异构数据融合的火灾现场信息分析研究[J]. 消防科学与技术, 2023, 42(6): 864-869.
	Duan Rufu, Tong Runze. Fire scene information analysis based on heterogeneous data fusion[J]. Fire Science and Technology, 2023, 42(6): 864-869.
[8]	Iftikhar S, Gill S S, Song Chenghao, et al. AI-based fog and edge computing: a systematic review, taxonomy and future directions[J]. Internet of Things, 2023, 21: DOI:10.1016/J.IOT.2022.100674.
[9]	Balasubramaniam S, Chirchi V, Kadry S, et al. The road ahead: emerging trends, unresolved issues, and concluding remarks in generative AI: a comprehensive review[J]. International Journal of Intelligent Systems, 2024: DOI:10.1155/2024/4013195.
[10]	Zhuang Licong, Zhong Xiaorong, Xu Linjie, et al. Visual SLAM for unmanned aerial vehicles: localization and perception[J]. Sensors, 2024, 24(10): DOI:10.3390/s24102980.
[11]	熊光明, 罗震, 孙冬, 等. 基于红外相机和毫米波雷达融合的烟雾遮挡无人驾驶车辆目标检测与跟踪[J]. 兵工学报, 2024, 45(3): 893-906. doi: 10.12382/bgxb.2022.0602
	Xiong Guangming, Luo Zhen, Sun Dong, et al. Object detection and tracking for unmanned vehicles based on fusion of infrared camera and mmw radar in smoke-obscured environment[J]. Acta Armamentarii, 2024, 45(3): 893-906. doi: 10.12382/bgxb.2022.0602
[12]	高寒, 于英, 李力, 等. 全球导航卫星系统拒止条件下无人机影像快速级联检索方法[J]. 光学学报, 2024, 44(24): 267-276.
	Gao Han, Yu Ying, Li Li, et al. Fast cascade retrieval method for UAV images with GNSS rejection[J]. Acta Optica Sinica, 2024, 44(24): 267-276.
[13]	周琪浩, 赵荣煊, 黄庆哲, 等. 消防救援无人化作战模式探索研究[J]. 中国应急救援, 2025(5): 20-27.
	Zhou Qihao, Zhao Rongxuan, Huang Qingzhe, et al. Exploration of research on unmanned operation modes in fire rescue[J]. China Emergency Rescue, 2025(5): 20-27.
[14]	NFPA 72-2025 National fire alarm & signaling code[S].
[15]	王秉, 王兵, 廖慧敏. 元宇宙安全的内涵及治理路径研究[J]. 中国安全科学学报, 2024, 34(2): 2-7. doi: 10.16265/j.cnki.issn1003-3033.2024.02.0921
	Wang Bing, Wang Bing, Liao Huimin. Research on connotation and governance path of metaverse safety & security[J]. China Safety Science Journal, 2024, 34(2): 2-7. doi: 10.16265/j.cnki.issn1003-3033.2024.02.0921
[16]	江田汉, 姜传胜, 高东风, 等. 安全风险“三道防线”系统防控模型研究:基于安全管理体系的视角[J]. 中国安全生产科学技术, 2025, 21(5): 28-35.
	Jiang Tianhan, Jiang Chuansheng, Gao Dongfeng, et al. Research on systematic prevention and control model of "three lines of defense" for safety risks: from perspective of safety management system[J]. Journal of Safety Science and Technology, 2025, 21(5): 28-35.
[17]	陈谋, 马浩翔, 雍可南, 等. 无人机安全飞行控制综述[J]. 机器人, 2023, 45(3): 345-366. doi: 10.13973/j.cnki.robot.220392
	Chen Mou, Ma Haoxiang, Yong Kenan, et al. Safety flight control of UAV: a survey[J]. Robot, 2023, 45(3): 345-366. doi: 10.13973/j.cnki.robot.220392
[18]	刘占省, 张安山, 王文思, 等. 数字孪生驱动的冬奥场馆消防安全动态疏散方法[J]. 同济大学学报(自然科学版), 2020, 48(7): 962-971.
	Liu Zhansheng, Zhang Anshan, Wang Wensi, et al. Dynamic fire evacuation guidance method for Winter Olympic venues based on digital twin-driven model[J]. Journal of Tongji University(Natural Science), 2020, 48(7): 962-971.
[19]	陈长坤, 张健, 焦伟冰, 等. 数字孪生消防救援技术系统架构及关键技术分析[J]. 中国安全科学学报, 2023, 33(8): 156-163. doi: 10.16265/j.cnki.issn1003-3033.2023.08.1363
	Chen Changkun, Zhang Jian, Jiao Weibing, et al. Research on architecture and key technologies of fire rescue technology system based on digital twin[J]. China Safety Science Journal, 2023, 33(8): 156-163. doi: 10.16265/j.cnki.issn1003-3033.2023.08.1363
[20]	Liu Yabo, Yang Jianhua, Zheng Yao, et al. Multi-robot coordination in complex environment with task and communication constraints[J]. International Journal of Advanced Robotic Systems, 2013, 10(1): 1-13. doi: 10.5772/52938
[21]	Korsah G A, Stentz A, Dias M B. A comprehensive taxonomy for multi-robot task allocation[J]. The International Journal of Robotics Research, 2013, 32(12): 1495-1512. doi: 10.1177/0278364913496484
[22]	Gielis J, Shankar A, Prorok A. A critical review of communications in multi-robot systems[J]. Current Robotics Reports, 2022, 3: 213-225. doi: 10.1007/s43154-022-00090-9
[23]	Heinonen H, Hämäläinen T, Koskela T, et al. Performance of sensor data process offloading on 5G-enabled UAVs[J]. Sensors, 2023, 23(2): DOI:10.3390/s23020864.
[24]	Ştefan A, Grigore L Ş, Molder C, et al. Research on heat transfer through a double-walled heat shield of a firefighting robot[J]. Machines, 2022, 10(10): DOI:10.3390/machines10100942.
[25]	Fujita J, Tamura Y, Amano H, et al. Novel exterior cover design for radiant heat resistance of firefighting robots in large-scale petrochemical complex fires[J]. ROBOMECH Journal, 2022, 9: DOI:10.1186/s40648-022-00229-5.
[26]	Li Sen, Yun Junying, Feng Chunyong, et al. An indoor autonomous inspection and firefighting robot based on SLAM and flame image recognition[J]. Fire, 2023, 6(3): DOI: 10.3390/fire6030093.

序号	关键词	突现强度	出现年份	结束年份	2016—2025年
1	消防	1.60	2016	2017	▃▃▂▂▂▂▂▂▂▂
2	免像控	3.93	2017	2019	▂▃▃▃▂▂▂▂▂▂
3	军事作战	2.36	2018	2019	▂▂▃▃▂▂▂▂▂▂
4	无人艇	3.99	2020	2021	▂▂▂▂▃▃▂▂▂▂
5	物联网	7.01	2021	2022	▂▂▂▂▂▃▃▂▂▂
6	资源配置	3.33	2021	2023	▂▂▂▂▂▃▃▃▂▂
7	俄乌冲突	1.52	2022	2023	▂▂▂▂▂▂▃▃▂▂
8	路径规划	4.34	2022	2023	▂▂▂▂▂▂▃▃▂▂
9	智慧农业	2.67	2022	2025	▂▂▂▂▂▂▃▃▃▃
10	医疗	5.45	2023	2025	▂▂▂▂▂▂▂▃▃▃

序号	关键词	突现强度	出现年份	结束年份	2016—2025年
1	火灾救援	2.63	2016	2021	▃▃▃▃▃▃▂▂▂▂
2	海上搜救	0.72	2019	2021	▂▂▂▃▃▃▂▂▂▂
3	云计算	1.22	2019	2021	▂▂▂▃▃▃▂▂▂▂
4	林火	2.36	2019	2021	▂▂▂▃▃▃▂▂▂▂
5	三维建模	0.9	2020	2021	▂▂▂▂▃▃▂▂▂▂
6	遥感	1.62	2020	2021	▂▂▂▂▃▃▂▂▂▂
7	监控	1.25	2020	2023	▂▂▂▂▃▃▃▃▂▂
8	石化装备	0.78	2021	2022	▂▂▂▂▂▃▃▂▂▂
9	城市消防	0.79	2022	2025	▂▂▂▂▂▂▃▃▃▃
10	消防通信	1.31	2022	2025	▂▂▂▂▂▂▃▃▃▃

评价指标	传统云— 端架构	“五层八步” 架构	优化幅度/%
任务保持率/%	8.2±2.6	97.2±1.4	+1085
态势更新延迟/ms	889.4±164.6	108.4±20.0	-88
平台安全保持率/%	34.7±7.6	69.5±3.4	+100
重调度时间/s	172.9±32.5	9.4±2.3	-95
任务冲突率/%	42.5±6.6	6.3±1.9	-85

面向复杂火灾场景的AI驱动无人化消防救援体系综述

A review of AI-driven unmanned rescue framework for complex fire scenarios

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 26

相关文章 15

编辑推荐

Metrics

本文评价

[1]	黄遥, 余嘉仪, 刘顶立, 熊柳源, 刘富斌, 唐加福. 寒区加油加气站消防救援可达性动态评估[J]. 中国安全科学学报, 2026, 36(6): 246-253.
[2]	贾晓芬, 赵玉晨, 赵佰亭, 胡锐, 梁镇洹. 数字孪生驱动的立井井筒结构性能在线监测[J]. 中国安全科学学报, 2026, 36(2): 66-76.
[3]	宫培松, 沈欣云, 韩博雯, 郭聖煜, 彭冠卿, 钟骏豪. 深基坑工程人员区域入侵行为预警数字孪生系统[J]. 中国安全科学学报, 2025, 35(8): 33-39.
[4]	陈涛, 张辉, 黄丽达, 田冉冉, 闫小丽. 应急指挥S3DA2框架与应急数字孪生战场关键技术[J]. 中国安全科学学报, 2025, 35(10): 205-212.
[5]	吴恙龙, 刘顶立, 袁狄平, 刘伟军, 徐志胜, 雷晓. 基于街区可达性的消防救援辖区划分方法[J]. 中国安全科学学报, 2025, 35(10): 181-189.
[6]	周洪涛, 赵开功, 张晓蕾, 孔言言, 王睿迪. 煤化工企业消防救援能力评估方法研究[J]. 中国安全科学学报, 2024, 34(S1): 205-211.
[7]	王晗. 基于数字孪生的钻机运行安全状态监测技术研究[J]. 中国安全科学学报, 2024, 34(S1): 134-139.
[8]	翟越, 雷尚学, 王轶宏, 王奥晨, 谢梓涵, 贾宇. 基于数字孪生的老旧建筑结构安全性定量预测方法[J]. 中国安全科学学报, 2024, 34(9): 69-77.
[9]	马树焕, 宁志强, 卫立新. 基于多核并行数字孪生的叉车稳定性预警方法及试验[J]. 中国安全科学学报, 2024, 34(8): 93-100.
[10]	马庆禄, 丁雪琴, 黄筱潇, 李君. 新型轨道消防救援车模块化系统设计[J]. 中国安全科学学报, 2024, 34(3): 70-75.
[11]	孙殿阁. 基于数字孪生技术的民用机场安全管理系统构建[J]. 中国安全科学学报, 2023, 33(S1): 222-227.
[12]	陈长坤, 张健, 焦伟冰, 陆童. 数字孪生消防救援技术系统架构及关键技术分析[J]. 中国安全科学学报, 2023, 33(8): 156-163.
[13]	潘卫军, 覃莉茹, 骆晨, 黄园晶. 通用航空森林消防救援数据适配性特征提取[J]. 中国安全科学学报, 2023, 33(11): 67-74.
[14]	黄萍, 张文龙, 叶圣琳, 余君, 余龙星. 基于TF-IDF和VOSviewer的我国应急救援现状可视化分析[J]. 中国安全科学学报, 2023, 33(11): 196-205.
[15]	鲁义, 杨帆, 刘欢, 陈健, 王金鹏, 李文辉. 气凝胶消防救援服内衬的热阻计算[J]. 中国安全科学学报, 2022, 32(1): 201-207.