近年来，随着机器人、人工智能等相关学科飞速崛起，煤矿原有的生产方式正在逐步向机器人化和智能化转变，为煤矿工业的发展带来新质生产力。煤矿井下移动机器人的研究对煤矿安全高效生产有着非常关键的现实意义，然而，煤矿井下环境复杂且相对封闭，在这样的工作环境下，地面常规使用的定位技术和路径规划方法无法直接应用。为此，本文设计煤矿井下移动机器人系统，针对该移动机器人的自主定位与路径规划技术进行研究，并研制出一台用于煤矿井下的移动机器人，旨在提高煤矿井下移动机器人的自主定位和路径规划能力。

针对煤矿井下封闭复杂的特殊环境，分析煤矿井下移动机器人的工作环境与功能需求，研制一台煤矿井下移动机器人。该机器人的硬件系统结构分为环境感知层、动作执行层和规划决策层。软件系统包括环境探测模块、定位模块、路径规划模块、通讯模块、运动控制模块和PC端模块，该机器人的研制为本文的自主定位与路径规划算法研究奠定基础。

针对煤矿井下移动机器人自主定位问题，提出利用机器人的行驶轨迹曲率和航向角来进行航位推算，而非传统的对复杂地面参数进行估计的方法，建立转弯时的煤矿井下移动机器人定位模型，避免传统机器人定位模型的复杂性。并提出一种卡尔曼滤波算法、LSTM(Long Short Term Memory)算法和自适应蒙特卡罗算法相结合的煤矿井下移动机器人自主定位方法。实验表明，该方法能够实现煤矿井下移动机器人的自主定位，机器人的平均位置误差为0.027m，平均航向角误差为1.74°。

针对煤矿井下移动机器人的全局路径规划问题，结合煤矿井下环境障碍物较多的特点，提出优化蚁群的全局路径规划算法。该算法首先通过改进随机比例规则函数，避免规划路径进入死锁点，其次，通过遗传算法对蚁群算法中的参数进行优化，通过连续的迭代过程筛选出最优参数组合。最后，为了使路径规划的结果更加平滑且符合机器人的实际运动轨迹，引入了Bezier曲线对所规划路径拐点进行平滑处理。实验表明，该算法在模拟煤矿井下障碍物较多环境中，路径长度由30.04m减少到28.62m，收敛次数由77次收敛变为11次收敛。

针对避障路径规划过程中机器人速度与安全性难以平衡的问题，提出参数自适应的DWA（Dynamic Window Approach）局部路径规划算法，该算法能够根据机器人与障碍物之间的距离动态调整速度权值，以适应环境变化。从而使机器人在避障过程中以保证安全的前提下，缩短整个路径的行驶时间。实验表明，该算法与高速度权值相比可以提高安全性，而与低速度权值相比行驶总时间降低34.25%。

在西安科技大学模拟煤矿巷道中进行现场实验，以研制的煤矿井下移动机器人为平台，验证本文所提出的自主定位与路径规划算法的可行性和可靠性，实验表明，本文提出的理论和方法为煤矿井下环境中的移动机器人自主定位与路径规划算法提供理论支撑，也可为机器人在其他相对封闭复杂环境下的导航算法研究奠定基础。

In recent years, with the rapid rise of disciplines such as robotics and artificial intelligence, the original production methods of coal mines are gradually shifting towards roboticization and intelligence, bringing new quality productivity to the development of coal mining industry. The research on mobile robots in coal mines has crucial practical significance for the safe and efficient production of coal mines. However, the underground environment of coal mines is complex and relatively closed, and in such a working environment, the positioning technology and path planning methods commonly used on the ground cannot be directly applied. Therefore, this article designs a coal mine underground mobile robot system, conducts research on the autonomous positioning and path planning technology of the mobile robot, and develops a mobile robot for coal mine underground, aiming to improve the autonomous positioning and path planning capabilities of the coal mine underground mobile robot.

In response to the complex and enclosed special environment of coal mines, the working environment and functional requirements of mobile robots in coal mines are analyzed, and a coal mine underground mobile robot is developed. The hardware system structure of the robot is divided into an environment perception layer, an action execution layer, and a planning and decision-making layer. The software system includes an environment detection module, a positioning module, a path planning module, a communication module, a motion control module, and a PC module. The development of this robot lays the foundation for the research of autonomous positioning and path planning algorithms in this paper.

Aiming at the problem of autonomous positioning of mobile robots in coal mines, a method is proposed to use the curvature and heading angle of the robot's travel trajectory for navigation calculation, instead of traditional methods for estimating complex ground parameters. A positioning model for mobile robots in coal mines during turns is established to avoid the complexity of traditional robot positioning models. And propose a coal mine underground mobile robot autonomous positioning method that combines Kalman filtering algorithm, LSTM (Long Short Term Memory) algorithm, and adaptive Monte Carlo algorithm. The experiment shows that this method can achieve autonomous positioning of mobile robots in coal mines underground, with an average position error of 0.027m and an average heading angle error of 1.74 °.

Aiming at the global path planning problem of mobile robots in coal mines, combined with the characteristics of many obstacles in the underground environment of coal mines, an optimized ant colony global path planning algorithm is proposed. The algorithm first improves the random proportion rule function to avoid the planned path from entering a deadlock point. Secondly, genetic algorithm is used to autonomously optimize the parameters in the ant colony algorithm. Finally, the Bezier curve was introduced to smooth the inflection points of the planned path. The experiment shows that the algorithm reduces the path length from 30.04m to 28.62m and the convergence frequency from 77 to 11 in simulating environments with many obstacles underground in coal mines.

A parameter adaptive DWA (Dynamic Window Approach) local path planning algorithm is proposed to address the issue of balancing robot speed and safety in obstacle avoidance path planning. This algorithm can dynamically adjust speed weights based on the distance between the robot and obstacles to adapt to environmental changes. Thus, the robot can shorten the travel time of the entire path while ensuring safety during obstacle avoidance. Experiments have shown that this algorithm can improve safety compared to high speed weights, while reducing total driving time by 34.25% compared to low speed weights.

An on-site experiment was conducted in a simulated coal mine tunnel at Xi'an University of Science and Technology, using the developed underground mobile robot as a platform to verify the feasibility and reliability of the proposed autonomous positioning and path planning algorithm. The experiment showed that the theory and method proposed in this paper provide theoretical support for the autonomous positioning and path planning algorithm of mobile robots in the underground environment of coal mines, and also lay a foundation for the research of navigation algorithms for robots in other relatively closed and complex environments.

[1] 国家发改委、国家能源局下发《能源技术革命创新行动计划(2016-2030年)》和《能源技术革命重点创新行动路线图》[J].电力与能源,2016,37(05):609.

[2] 胡庆明.《能源技术革命重点创新行动路线图发布》[J].石油石化节能,2016,6(06):7.

[3] Zhang K F,Zhu M,Wang Y J,et al. Underground mining intelligent response and rescue systems[J]. Procedia Earth & Planetary Science,2009,1(1):1044-1053.

[4] 戚兰丽,常志亮,张进领.论煤矿灾害防治技术现状与发展[J].中国科技博览,2013(20):545-545.

[5] 袁浩钦,秦江涛.当前我国煤矿灾害防治存在的主要技术难题及应对策略[J].中国科技博览,2015(20):77-77.

[6] 王国庆,许红盛,王恺睿.煤矿机器人研究现状与发展趋势[J].煤炭科学技术,2014,42(2):73-77.

[7] 由韶泽,朱华,赵勇,等.煤矿救灾机器人研究现状及发展方向[J].工矿自动化,2017,43(4):14-18.

[8] 孙继平.煤矿信息化自动化新技术与发展[J].煤炭科学技术,2016,44(1):19-23.

[9] Khaireddin A,RTalib M S,Haron H. Review on simultaneous localization and mapping(SLAM)[C]. IEEE International Conference on Control System,Computing and Engineering. IEEE,2015:85-90.

[10] 步新宇,项玉.智能移动机器人定位技术[J].科技展望,2016,26(10).

[11] Leonard J J,Durrant Whyte H F,Cox I J. Dynamic map building for an autonomous mobile robot[J]. International Journal of Robotics Research,1992,11(4):286-298.

[12] 冯子龙,刘健,刘开周.AUV自主导航航位推算算法的研究[J].机器人,2005,(2):168-172.

[13] 陈磊,梁强.GPS原理及应用简介[J].科技信息(学术研究),2008,(22):188-190.

[14] 张璞.煤矿井下移动机器人自主定位方法研究[D].西安科技大学,2019.

[15] 戚鸿飞,孙海淇.基于捷联惯性导航的井下机车精确定位系统[J].煤矿机电,2011,(04):62-65.

[16] 李群明,熊蓉,褚健.室内自主移动机器人定位方法研究综述[J].机器人,2003,(06):560-567+573.

[17] 张梦轩,苏治宝,索旭东.移动机器人定位方法研究综述[J].车辆与动力技术,2023,(04):56-62.

[18] 郁露,唐超礼,黄友锐,等.基于UWB和IMU的煤矿机器人紧组合定位方法研究[J].工矿自动化,2022,48(12):79-85.

[19] 左洋,李伟宏.UWB精准定位技术在电机车无人驾驶系统中的应用[J].工矿自动化,2023,49(S1):47-49.

[20] 李海丰,胡遵河,陈新伟.PLP-SLAM:基于点、线、面特征融合的视觉SLAM方法[J].机器人,2017,39(02):214-220+229.

[21] Hongming S,Qu Z,Hanchen LU,et al. A distributed approach for lidar-based relative state estimation of multi-UAV in GPS-denied environments[J]. Chinese Journal of Aeronautics,2022,35(1):59-69.

[22] 司垒,王忠宾,魏东,等.基于IMU-LiDAR紧耦合的煤矿防冲钻孔机器人定位导航方法[J/OL].煤炭学报,2024.

[23] 杨林,马宏伟,王岩.基于激光惯性融合的煤矿井下移动机器人SLAM算法[J].煤炭学报,2022,47(09):3523-3534.

[24] 韩震峰.面向煤矿井下探测的多节履带式机器人及其关键技术研究[D].哈尔滨:哈尔滨工业大学,2012.

[25] 杜钊君,吴怀宇.基于激光测距与双目视觉信息融合的移动机器人SLAM研究[J].计算机测量与控制,2013,21(1):180-183.

[26] 韩冬桂.搜索机器人定位技术研究[D].武汉:华中科技大学,2009.

[27] 刘洞波.移动机器人粒子滤波定位与地图创建方法研究[D].长沙:湖南大学,2013.

[28] Li, Mingyang,Mourikis,Anastasios I. High-precision，consistent EKF-based visual-inertial odometry[J]. International Journal of Robotics Research,2013,32(6):690-711.

[29] 杨海,李威,罗成名,等.基于捷联惯导的采煤机定位定姿技术实验研究[J].煤炭学报,2014,39(12):2550-2556.

[30] 鱼瑜,吴明亮,张来喜等.多传感器融合定位方法探讨[J].导航定位学报,2023,11(05):151-163.

[31] 高金波,田骞,周焘.煤矿智能巡检机器人定位技术研究[J].煤炭技术,2023,42(11):234-236.

[32] 梁春苗,姚宁平,姚亚峰,等.煤矿用钻孔机器人钻臂定位误差补偿研究[J].煤田地质与勘探,2024,52(03):153-163.

[33] Zhan W W,Wang W,Chen,N C,et al. Path planning strategies for UAV based on improved A*algorithm[J]. Geomatics and Information Science of Wuhan University, 2015,40(3):315-320.

[34] 孙炜,吕云峰,唐宏伟,等.基于一种改进A*算法的移动机器人路径规划[J].湖南大学学报(自然科学版),2017,44(4): 94-101.

[35] 陈秋莲,蒋环宇,郑以君.机器人路径规划的快速扩展随机树算法综述[J].计算机工程与应用,2019,55(16):10-17.

[36] Huang H C. FPGA-based parallel metaheuristic PSO algorithm and its application to global path planning for autonomous robot navigation[J]. Journal of Intelligent and Robotic Systems,2014,76(3-4):475-488.

[37] 陈雄,赵一路,韩建达,等.一种改进的机器人路径规划的蚁群算法[J].控制理论与应用, 2010,27(6):821-825.

[38] 徐菱,付文浩,江文辉,等.基于16方向24邻域改进蚁群算法的移动机器人路径规划[J].控制与决策,2021,36(5):1137-1146.

[39] Arora T,Gigras Y,Arora V. Robotic path planning using genetic algorithm in dynamic environment[J]. International Journal of Computer Applications,2014,89(11):8-12.

[40] Nazarahari M,Khanmirza E,Doostie S. Multi-objective multi-robot path planning in continuous environment using an enhanced genetic algorithm[J]. Expert Systems with Applications,2019,115:106-120.

[41] 禹建丽,V.K roumov,孙增圻,等.一种快速神经网络路径规划算法[J].机器人,2001,(03):201-205.

[42] Peng J, Huang Y, Luo G. Robot path planning based on improved A* algorithm[J]. Cybemetics and Information Technologies,2015,15(2):171-180.

[43] Shang E,Dai B,Nie Y,et al. An improved A-Star based path planning algorithm for autonomous land vehicles[J]. International Journal of Advanced Robotic Systems, 2020,17(5):1-13.

[44] Leng Z,Dong M Z,Dong G Q,et al. A rapid path planner for autonomous ground vehicle using section collision detection[J]. Journal of Shanghai Jiaotong University (Science),2009,14(3):306-309.

[45] Qi J,Yang H,Sun H.MOD-RRT*: a sampling-based algorithm for robot path planning in dynamic environment[J]. IEEE Transactions on Industrial Electronics,2020,68(8):7244-7251.

[46] 朱子祺,李创业,代伟.基于G-RRT~*算法的煤矸石分拣机器人路径规划[J].工矿自动化,2022,48(03):55-62.

[47] Lazarowska A. Multi-criteria ACO-based algorithm for ship's trajectory planning[J]. Transnav International Journal on Marine Navigation and Safety of Sea Transportation,2017,11(1):31-36.

[48] Lee J. Heterogeneous-ants-based path planner for global path planning of mobile robot applications[J]. International Journal of Control Automation and Systems,2017,15(5):1-16.

[49] SONG B,MIAO H,XU L. Path planning for coal mine robot via improved ant colony optimization algorithm [J]. Systems Science & Control Engineering,2021,9(1):283-9.

[50] 于进,王建鹏,张海君.栅格环境地图启发下煤矿井环境管路安装机器人协同控制[J].自动化与仪表,2023,38(12):38-42.

[51] Wang X,Shi Y,Ding D,et al. Double global optimum genetic algorithm-particle swarm optimization-based welding robot path planning[J]. Engineering Optimization,2016,48(2): 299-316.

[52] Nobile M S,Cazzaniga P,Besozzi D,et al. Fuzzy self-tuning PSO: a settings-free algorithm for global optimization[J]. Swarm and Evolutionary Computation,2017,11(8):1251-1255.

[53] 赵少林,程杰.基于粒子群并行优化的煤矿井下机器人路径规划[J].计算机测量与控制,2014,22(05):1600-1602+1615.

[54] Liu E,Yao X. AGV path planning based on improved genetic algorithm and implementation platform[J]. Computer Integrated Manufacturing Systems, 2017,23(3):465-472.

[55] Cao Y,Wei w Y,Bai Y,et al. Multi-base multi-UAV cooperative reconnaissance path planning with genetic algorithm[J]. Cluster Computing,2019,22(3):5175-5184.

[56] 周巍,李元宗.基于改进遗传算法的煤矿探测机器人路径规划[J].太原理工大学学报,2010,41(04):364-367.

[57] Khelchandra T,Huang J,Debnath S. Path planning of mobile robot with neuro-genetic-fuzzy technique in static environment[J]. International Journal of Hybrid Intelligent Systems,2014,11(2):71-80.

[58] Bayerlein H,Theile M,Caccamo M,et al. Multi-UAV path planning for wireless data harvesting with deep reinforcement learning[J]. IEEE Open Journal of the Communications Society,2021,2:1171-1187.

[59] Qie H,Shi D X,Shen T L,et al. Joint optimization of multi-UAV target assignment and path planning based on multi-agent reinforcement learning[J]. IEEE Access,2019,7:146264-146272.

[60] 唐立伟.基于模糊神经网络的煤矿救援机器人路径规划研究[J].现代制造技术与装备,2020,56(11):71-73.

[61] 白云.煤矿蛇形探测机器人位姿控制方法研究[D].西安科技大学,2019.

[62] 李红岩.煤矿蛇形搜寻机器人路径规划策略研究[D].西安科技大学,2018.

[63] 黄玲,胡蔚蔚.基于改进蚁群算法的果蔬采摘机器人三维路径规划[J].农机化研究,2016,38(9):38-42.

[64] 赵娟平.移动机器人路径规划的蚁群优化算法研究[D].沈阳:东北大学,2012.

[65] 李晓静,余东满.煤炭勘探及救援机器人最优路径规划研究[J].工矿自动 化,2017,43(3):24-29.

[66] 徐岩,崔媛媛.基于Q-IGA动态拟合贝塞尔曲线的路径规划[J].湖南大学学报(自然科学版).2020,47(10):68-75.