面向消防自主作业的灭火射流运动学控制研究

doi:10.3901/JME.2025.10.450

摘要/Abstract

摘要： 面向消防灭火自主作业，提出一种基于运动学模型的灭火射流控制方法。将射流连续体离散成关节变量，推导射流运动学模型，针对建模过程中多因素强耦合造成的非线性关系难以显式表征问题，建立基于自回归积分滑动平均模型(Autoregre ssive integratal moving average model,ARIMA)和长短时记忆网络(Long short-term memory,LSTM)的灭火射流轨迹预测模型。考虑实际作业过程中的障碍物干扰问题，重构烟花算法的初始种群分布和下一代种群确定方法，实现有障碍物约束条件下的射流驱动关节运动学逆解。建立考虑死区特性和喷射反力的关节动力学模型，设计基于径向基网络观测器的角度跟踪滑模控制器，实现对期望角度的跟踪控制，并在实际灭火机器人上开展验证试验。结果表明，灭火射流轨迹模型预测精度均优于LSTM、ARIMA及显式模型，关节角度跟踪控制算法的控制精度高于现有PD、模糊控制以及线性自抗扰控制器，在自主灭火作业测试中，所提方法的火点打击精度优于现有方法。

关键词: 消防自主作业, 灭火射流运动学模型, 灭火射流轨迹模型, 灭火射流控制, 时序网络预测

Abstract: A kinematic model-based fire extinguishing jet control method is proposed for autonomous firefightin. The jet continuum is discretized into joint variables, and a jet kinematic model is derived. To address the problem of nonlinear relationships caused by strong coupling of multiple factors in the modeling process of fire extinguishing jet, a trajectory prediction model based on ARIMA and LSTM is established. Considering the problem of obstacle interference in the actual process, the initial population distribution and next generation population determination method of the fireworks algorithm are reconstructed, and the inverse kinematics solution of the jet driven joint under obstacle constraints is achieved. A joint dynamics model considering dead zone characteristics and injection reaction force is established, and an angle tracking sliding mode controller based on radial basis network observer is designed to achieve robust tracking control of the desired angle. Verification experiments are conducted on actual fire extinguishing robots. The results show that the prediction accuracy of the fire jet trajectory model is superior to LSTM, ARIMA, and explicit models. The control accuracy of the joint angle tracking control algorithm is higher than that of existing PD, fuzzy control, and linear self disturbance rejection controllers. In the testing of autonomous fire extinguishing operations, the proposed method has better fire point strike accuracy than existing methods.

Key words: autonomous firefighting operations, kinematic model of fire extinguishing jet, trajectory model of fire extinguishing jet, control of fire extinguishing jet, ARIMA-LSTM

中图分类号:

TP273

潘禄, 朱劲松, 陈正升, 刘中冠, 赵桔贤, 李伟. 面向消防自主作业的灭火射流运动学控制研究[J]. 机械工程学报, 2025, 61(10): 450-463.

PAN Lu, ZHU Jinsong, CHEN Zhengsheng, LIU Zhongguan, ZHAO Juxian, LI Wei. Research on the Kinematic Control of Fire Extinguishing Jet for Autonomous Firefighting[J]. Journal of Mechanical Engineering, 2025, 61(10): 450-463.

参考文献

[1] 李敏，林志军，王德明，等. 我国煤矿重特大火灾事故统计分析[J]. 中国安全科学学报，2023，33(1)：115-121. LI Min，LIN Zhijun，WANG Deming，et al. Statistical analysis of major coal mine fire accidents in China[J]. Chinese Journal of Safety Sciences，2023，33(1)：115-121.
[2] CHAUMETTE F，HUTCHINSON S. Visual servo control. I. Basic approaches[J]. IEEE Robotics & Automation Magazine，2006，13(4)：82-90.
[3] KIM J，STARR J，LATTIMER B. Firefighting robot stereo infrared vision and radar sensor fusion for imaging through smoke[J]. Fire Technology，2015，51(4)：823-845.
[4] MIYASHITA T，SUGAWA O，IMAMUR T，et al. Modeling and analysis of water discharge trajectory with large capacity monitor[J]. Fire Safety Journal，2014，63(2014)：1-8.
[5] ZHU J，LI W，LIN D，et al. Study on water jet trajectory model of fire monitor based on simulation and experiment[J]. Fire Technology，2019，55(3)：773-787.
[6] LIN Y，JI W，HE H，et al. Two-stage water jet landing point prediction model for intelligent water shooting robot[J]. Sensors，2021，21(8)：2704.
[7] 闵永林，陈晓阳，陈池，等. 考虑俯仰角的消防水炮射流轨迹理论模型[J]. 机械工程学报，2011，47(11)：134-138.MIN Yonglin，CHEN Xiaoyang，CHEN Chi，et al. Theoretical model of fire water cannon jet trajectory considering pitch angle[J]. Journal of Mechanical Engineering，2011，47(11)：134-138.
[8] QU Y，LIU X，ZHANG M，et al. Analysis and evaluation of drop point for water jet based on wave model[J]. Journal of Engineering Research，2021，9(1)：229-246.
[9] ZHU J，LI W，LIN D，et al. Intelligent fire monitor for fire robot based on infrared image feedback control[J]. Fire Technology，2020，56(5)：2089-2109.
[10] MCNEIL J，LATTIMER B. Robotic fire suppression through autonomous feedback control[J]. Fire Technology，2017，53(3)：1171-1199.
[11] 林达. 城市主战消防车自动消防炮研究[D]. 徐州：中国矿业大学，2020.LIN Da. Research on automatic fire monitor for urban main battle fire vehicle[D]. Xuzhou：China University of Mining and Technology，2020.
[12] SUN J，LI W，HE M. Analysis of fire water monitor jet reaction forces and their influences on the roll stabilities of urban firefighting vehicle[J]. Fire Technology，2019，55(6)：2547-2566.
[13] 胡海兵，崔世林，张波，等. 基于线性自抗扰的消防水炮控制系统[J]. 现代电子技术，2019，42(5)：183-186.HU Haibing，CUI Shilin，ZHANG Bo，et al. A fire water monitor control system based on linear autodisturbance rejection[J]. Modern Electronic Technology，2019，42(5)：183-186.
[14] 王久健，杨绍普，刘永强，等. 一种基于空间卷积长短时记忆神经网络的轴承剩余寿命预测方法[J]. 机械工程学报，2021，57(21)：88-95.WANG Jiujian，YANG Shaopu，LIU Yongqiang，et al. A method of bearing remaining useful life estimation based on convolutional long short-term memory neural network[J]. Journal of Mechanical Engineering，2021，57(21)：88-95.
[15] WU Peihao，LÜ Nawei，SONG Yuhang，et al. Li-ion battery failure warning methods for energy-storage systems[J]. Chinese Journal of Electrical Engineering，2024，10(1)：86-100.
[16] 张朝龙，赵筛筛，何怡刚. 基于信息熵与PSO-LSTM的锂电池组健康状态估计方法[J]. 机械工程学报，2022，58(10)：180-190.ZHANG Chaolong，ZHAO Shaishai，HE Yigang. A health state estimation method for lithium battery pack based on information entropy and PSO-LSTM[J]. Journal of Mechanical Engineering，2022，58(10)：180-190.
[17] 于仲安，张军令，陈可怡. 基于GA-ELM的锂电池SOC估计及主动均衡[J]. 电气工程学报，2024，19(1)：326-333.YU Zhongan，ZHANG Junling，CHEN Keyi. SOC Estimation and active equalization of lithium battery based on GA-ELM[J]. Journal of Electrical Engineering，2024，19(1)：326-333.
[18] LI Kuijie，LI Yalun，RUI Xinyu，et al. Experimental study on the effect of state of charge on failure propagation characteristics within battery modules[J]. Chinese Journal of Electrical Engineering，2023，9(3)：3-14.
[19] 任济寰，吴祥，薄煜明，等. 基于增强上下文信息长短期记忆网络的弹道轨迹预测[J]. 兵工学报，2023，44(2)： 462-471.REN Jihuan，WU Xiang，BO Yuming，et al. Ballistic trajectory prediction based on enhanced contextual information and short-term memory network[J]. Journal of Ordnance Engineering，2023，44(2)：462-471.
[20] LI C，CHIANG T W. Complex neuro fuzzy ARIMA forecasting-a new approach using complex fuzzy sets[J]. IEEE Transactions on Fuzzy Systems，2013，21(3)：567-584.
[21] 王储，袁晓明，杨志刚，等. 消防水炮射流轨迹理论模型研究[J]. 燕山大学学报，2020，44(5)：442-449. WANG Chu，YUAN Xiaoming，YANG Zhigang，et al. Research on the theoretical model of fire water cannon jet trajectory[J]. Journal of Yanshan University，2020，44(5)：442-449.
[22] 董静怡，庞景月，彭宇，等. 集成LSTM的航天器遥测数据异常检测方法[J]. 仪器仪表学报，2019，40(7)：22-29. DONG Jingyi，PANG Jingyue，PENG Yu，et al. An anomaly detection method for spacecraft telemetry data integrated with LSTM[J]. Journal of Instrumentation，2019，40(7)：22-29.
[23] 张启航，邵敏，任树雄，等. 仿象鼻气动连续体机器人的运动学建模与运动控制[J]. 西安交通大学学报，2021，55(8)：25-32. ZHANG Qihang，SHAO Min，REN Shuxiong，et al. Kinematic modeling and motion control of an elephant trunk like pneumatic continuum robot[J]. Journal of Xi’an Jiaotong University，2021，55(8)：25-32.
[24] 朱劲松. 基于机器视觉的智能化消防水炮控制研究[D]. 徐州：中国矿业大学，2020. ZHU Jinsong. Research on intelligent fire water monitor control based on machine vision[D]. Xuzhou：China University of Mining and Technology，2020.
[25] 孙凤山，范孟豹，曹丙花，等. 基于混沌映射与差分进化自适应教与学优化算法的太赫兹图像增强模型[J]. 仪器仪表学报，2021，42(4)：92-101. SUN Fengshan，FAN Mengbao，CAO Binghua，et al. A terahertz image enhancement model based on chaotic mapping and differential evolution adaptive teaching and learning optimization algorithm[J]. Journal of Instrumentation，2021，42(4)：92-101.
[26] DIANOV A，ANUCHIN A. Phase loss detection using voltage signals and motor models：A review[J]. IEEE Sensors Journal，2021，21(23)：26488-26502.
[27] 赵健，邓志辉，朱冰，等. 基于RBF网络滑模的电动助力制动系统液压力控制[J]. 机械工程学报，2020，56(24)：106-114. ZHAO Jian，DENG Zhihui，ZHU Bing，et al. Hydraulic pressure control of electric power assisted braking system based on RBF network sliding mode[J]. Journal of Mechanical Engineering，2020，56(24)：106-114.
[28] HARTMAN E，KEELER J，KOWALSKI J. Layered neural networks with Gaussian hidden units as universal approximations[J]. Neural Computation，1990，3(2)：210-215.