随着矿山智能化开采向深部推进以及无人驾驶技术的持续进步，矿用无人驾驶系统正逐步实现商业化应用。实际矿区作业条件复杂多变，需要应对多种严峻挑战，对算法适用性和路径精准规划性提出严格要求。露天矿无人驾驶路径规划功能的实现依赖于动态障碍物检测、地图构建及高效的路径规划算法。本文针对露天矿无人驾驶卡车的路径规划与智能绕障问题，以矿用卡车搭载激光雷达为研究对象，对道路运输最优路径进行规划研究，旨在提升全局路径规划的效率与质量，动态障碍物检测与跟踪的准确性、以及局部路径规划的实时避障能力。本文的主要工作内容如下：

（1）矿用卡车全局最优路径的模型规划。针对全局路径规划过程中存在的路径质量不足、计算效率低、安全性差等问题，提出SPO-Hybrid A*算法。通过优化节点拓展方式、去除冗余节点、设计新的启发函数以及对路径进行平滑处理，显著提高算法的计算效率和路径质量，实验结果表明：该算法路径搜索时间降低10.26%，路径搜索节点减少30.57%，能够更好地适应复杂矿区环境。

（2）运输道路环境中动态障碍物检测及跟踪。露天矿非结构化道路下障碍物检测存在漏检误检、环境适应性差以及轨迹预测精度欠佳等问题，提出自适应聚类和多特征数据关联的MF-DBSCAN算法。该算法通过多特征加权和关联分析，更准确地描述数据点之间的关系，实验结果表明：目标检测精度提升16.29%，识别范围提升41.5%增强了对噪声和异常点的鲁棒性。

（3）建立矿用卡车局部避障的动态最优模型。针对全局路径规划算法在实时避障方面存在的局限性，无法及时应对突然出现的动态障碍物，易陷入局部最优等问题，提出基于自适应DWA的实时避障规划方法。通过动态调整窗口大小、优化评价函数及增强姿态调整能力，显著提升算法的实时性、安全性和适应性。实验结果表明：自适应DWA算法避障平均时间降低24.36%，能够根据环境复杂度动态调整搜索空间，生成更平滑、更安全的路径，并有效应对动态障碍物的干扰。

本研究通过融合三种算法的优势，创新性地提出一种混合算法进行无人驾驶矿用卡车路径规划，实现全局路径高效规划与局部避障实时响应的协同优化。实验验证表明，该方法在复杂矿区环境下显著提升了路径规划效率和行驶安全性，为露天矿无人驾驶系统提供了可靠的技术支撑。

With the advancement of intelligent mining to the depth of mines and the continuous progress of unmanned technology, mining unmanned systems are gradually realising commercial applications. The actual operating conditions in the mining area are complex and changeable, and need to cope with a variety of severe challenges, which puts strict requirements on the applicability of algorithms and the accurate planning of paths. The realisation of driverless path planning in open pit mines relies on dynamic obstacle detection, map construction and efficient path planning algorithms. In this paper, we focus on the path planning and intelligent obstacle avoidance for driverless trucks in open pit mines, and take mining trucks equipped with LiDAR as the research object to plan the optimal paths for road transport, aiming to improve the efficiency and quality of global path planning, the accuracy of dynamic obstacle detection and tracking, and real-time obstacle avoidance capability of local path planning. The main work of this paper is as follows:

(1) Model planning of global optimal path for mining trucks. Aiming at the problems of insufficient path quality, low computational efficiency, and poor security in the global path planning process, the SPO-Hybrid A* algorithm is proposed. By optimising the node expansion method, removing redundant nodes, designing a new heuristic function and smoothing the path, the computational efficiency and path quality of the algorithm are significantly improved. Experimental results show that the algorithm's path search time is reduced by 10.26%, and the number of path search nodes is reduced by 30.57%, and it can be better adapted to the complex mining environment.

(2) Dynamic obstacle detection and tracking in transport road environment. Obstacle detection under unstructured roads in open pit mines has problems such as omission and misdetection, poor environmental adaptability and poor trajectory prediction accuracy, etc. The MF-DBSCAN algorithm with adaptive clustering and multi-feature data association is proposed. The algorithm describes the relationship between data points more accurately through multi-feature weighting and correlation analysis, and the experimental results show that: the target detection accuracy is improved by 16.29%, and the recognition range is improved by 41.5% to enhance the robustness to noise and anomalies.

(3) Establish the dynamic optimal model of local obstacle avoidance for mining trucks. Aiming at the limitations of global path planning algorithms in real-time obstacle avoidance, which can't respond to sudden dynamic obstacles in a timely manner, and is easy to fall into the local optimum and other problems, the real-time obstacle avoidance planning method based on adaptive DWA is proposed. By dynamically adjusting the window size, optimising the evaluation function and enhancing the attitude adjustment ability, the real-time, safety and adaptability of the algorithm are significantly improved. Experiments reveal that the adaptive DWA algorithm lowers the average obstacle avoidance time by 24.36%, can dynamically adjust the search space according to the environmental complexity, generates smoother and safer paths, and effectively copes with the interference of dynamic obstacles.

In this study, by integrating the advantages of the three algorithms, a hybrid algorithm is innovatively proposed for unmanned mining truck path planning, which achieves the synergistic optimisation of efficient global path planning and local obstacle avoidance real-time response. Experimental validation shows that the method significantly improves the path planning efficiency and driving safety under the complex mining environment, providing a reliable technical support for the driverless system in open pit mines.

丁震, 孟峰. 矿用无人卡车国内外研究现状及关键技术[J]. 中国煤炭, 2020, 46(2): 42-49.

李明杰, 刘小飞. 基于毫米波传感器与激光雷达信号融合的自动驾驶障碍物感知方法[J]. 激光杂志, 2023, 44(06): 199-203.

Adenso-Diaz B, Lev B, Artime R. Simulation in dynamic environments: optimization of transportation inside a coal mine [J]. IIE Transactions, 2004, 36(6): 547-555.

史岩岩. 矿用卡车无人驾驶感知技术的设计与应用[J]. 工矿自动化, 2022, 48(S2): 94-97.

李庆玲, 张慧祥, 赵旭阳, 等. 露天矿无人驾驶自卸卡车发展综述[J]. 煤炭工程, 2021, 53(02): 29-34.

Huang Q, Zheng G. Route Optimization for Autonomous Container Truck Based on Rolling Window[J]. International Journal of Advanced Robotic Systems, 2016, 13(3): 112-112.

SungKeun K, JungWoo J, S W N. Automatic Creation of Heuristic-Based Truck Movement Paths for Construction Equipment Control[J]. Applied Sciences, 2021, 11(13): 5837-5837.

Xiaofeng W, Fei L, Sheng Z, et al. Obstacle Avoidance Path Planning Strategy for Autonomous Vehicles Based on Genetic Algorithm [J]. Promet – Traffic & Transportation, 2024, 36(4): 733-748.

张博强, 张成龙, 冯天培, 等. 基于JPS和变半径RS曲线的Hybrid A*路径规划算法[J]. 郑州大学学报(工学版), 2025, 46(02): 19-25.

柴森霖, 白润才, 刘光伟, 等. 基于改进遗传算法的露天矿运输路径优化[J]. 重庆大学学报, 2018, 41(02): 87-95.

Tianchuang M, Tianhong Y, Jin H, et al. Improved Hybrid A-Star Algorithm for Path Planning in Autonomous Parking System Based on Multi-Stage Dynamic Optimization[J]. International Journal of Automotive Technology, 2023, 24(2): 459-468.

谢春丽, 高胜寒, 孙学志. 融合改进A*算法和贝塞尔曲线优化的路径规划算法[J]. 重庆理工大学学报(自然科学), 2022, 36(07): 177-187.

Mei L D. Path Planning for Batch Picking of Warehousing and Logistics Robots Based on Modified A* Algorithm[J]. International Journal of Online Engineering (iJOE), 2018, 14 (11): 176-176.

Bhadoria A, Marwaha S. A chaotic hybrid optimization technique for solution of dynamic generation scheduling problem considering effect of renewable energy sources[J]. MRS Energy & Sustainability, 2022, 10(1): 52-93.

郭聚刚, 于军琪, 冯春勇, 等. 基于改进A*算法的机器人不平坦地形全局路径规划[J]. 计算机工程与应用, 2025, 61(05): 309-322.

Zhang H, Li M, Yang L. Safe Path Planning of Mobile Robot Based on Improved A* Algorithm in Complex Terrains[J]. Algorithms, 2018, 11(4): 44-44.

Jiang X, Wang Z, Dong C. A Path Planning Algorithm Based on Improved RRT Sampling Region[J]. Computers, Materials & Continua, 2024, 80(3): 4303-4323.

崔锡杰, 王晓军, 李晓航. 改进RRT算法的机器人全局路径规划[J]. 计算机工程与应用, 2025, 61 (04): 331-338.

王海芳, 张瑶, 朱亚锟, 等. 基于改进双向RRT*的移动机器人路径规划算法[J]. 东北大学学报(自然科学版), 2021, 42(08): 1065-1070+1142.

辛钢, 宋少忠, 张慧, 等. 基于改进遗传算法的路径规划问题应用[J]. 吉林大学学报(信息科学版), 2022, 40(06): 946-953.

Han J, Li W, Xia W, et al. Research on Complete Coverage Path Planning of Agricultural Robots Based on Markov Chain Improved Genetic Algorithm[J]. Applied Sciences, 2024, 14(21): 9868-9868.

Wang X, Ke F, Lu J. Research on AGV task path planning based on improved A* algorithm [J]. Virtual Reality & Intelligent Hardware, 2023, 5(3): 249-265.

Dejin D, Shiyin D, Lulu Z, et al. Path Planning Based on A-Star and Dynamic Window Approach Algorithm for Wild Environment[J]. Journal of Shanghai Jiaotong University (Science), 2024, 29(4): 725-736.

李忠林, 贾玉婷, 蒋晓丽. 改进遗传算法的移动机器人路径规划研究[J]. 现代信息科技, 2024, 8(23): 180-183+188.

Maria C, Burak K, Yu Z, et al. Deep Learning-Based Obstacle Detection and Classification With Portable Uncalibrated Patterned Light[J]. IEEE Sensors Journal, 2018, 18(20): 8416-8425.

Han J, Kim D, Lee M, et al. Road boundary detection and tracking for structured and unstructured roads using a 2D lidar sensor[J]. International Journal of Automotive Technology, 2014, 15(4): 611-623.

Gupta A, Jain S, Choudhary P, et al. Dynamic object detection using sparse LiDAR data for autonomous machine driving and road safety applications[J]. Expert Systems With Applications, 2024, 255(PB): 124636-124636.

Chen G, Yi H, Mao Z. 3D object detection based on image and LIDAR fusion for autonomous driving[J]. International Journal of Vehicle Information and Communication Systems, 2023, 8(3): 237-251.

Mushtaq H, Deng X, Azhar F, et al. PLC-Fusion: Perspective-Based Hierarchical and Deep LiDAR Camera Fusion for 3D Object Detection in Autonomous Vehicles[J]. Information, 2024, 15(11): 739-739.

Ren S, He K, Girshick R, et al. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017, 39(6): 1137-1149.

Mandouh A A, Ali O N E M, Mohamed A H M, et al. Automatic Building Extraction from Multispectral LiDAR Using Novel 3D Spatial Indices and Deep Learning Point CNN[J]. Journal of the Indian Society of Remote Sensing, 2024, 52(10): 2267-2280.

Yue Y, Cai Y, Wang D. GridNet-3D: A Novel Real‐Time 3D Object Detection Algorithm Based on Point Cloud[J]. Chinese Journal of Electronics, 2021, 30(5): 931-939.

Xu H, Yang L, Zhao S, et al. SPS-RCNN: Semantic-Guided Proposal Sampling for 3D Object Detection from LiDAR Point Clouds[J]. Sensors, 2025, 25(4): 1064-1064.

Li J, Dong S, Ding L. et al. MsSVT++: Mixed-Scale Sparse Voxel Transformer With Center Voting for 3D Object Detection[J]. IEEE transactions on pattern analysis and machine intelligence, 2024, 46(5): 3736-3752.

J S, Kuriakose A. An Optimized Framework for Detection and Tracking of Video Objects in Challenging Backgrounds[J]. The International journal of Multimedia & Its Applications, 2014, 6(4): 27-38.

Cretu A, Chagnon-Forget M, Payeur P. Selectively densified 3D object modeling based on regions of interest detection using neural gas networks[J]. Soft Computing, 2017, 21(18): 5443-5455.

Fan X, Xiao D, Li Q, et al. Snow-CLOCs: Camera-LiDAR Object Candidate Fusion for 3D Object Detection in Snowy Conditions[J]. Sensors, 2024, 24(13): 4158-4158.

陈乾. 动态环境下实时性混合路径规划算法研究[D]. 中国民用航空飞行学院, 2023.

任祥瑞, 王正刚, 汤俊杨. 改进JPS算法融合DWA的多机器人路径规划[J]. 计算机应用研究, 2024, 41(11): 3251-3257.

Filaretov F V, Yukhimets A D. Planning smooth paths for mobile robots in an unknown environment[J]. Journal of Computer and Systems Sciences International, 2017, 56(4): 738-748.

安林芳, 陈涛, 成艾国, 等. 基于人工势场算法的智能车辆路径规划仿真[J]. 汽车工程, 2017, 39(12): 1451-1456.

Giseo P, Mooryong C. Optimal Path Planning for Autonomous Vehicles Using Artificial Potential Field Algorithm[J]. International Journal of Automotive Technology, 2023, 24(5): 1259-1267.

MishraB, SevilE H. Path planning algorithm design using particle swarms optimization and artificial potential fields[J]. Electronics Letters, 2024, 60(18): e70038-e70038.

Tirumalapudi R, Vedaraj R S I. LiDAR Sensor Perception System for Self-Driving Cars[J]. International Journal of Vehicle Structures & Systems, 2023, 15(4): 555-558.

冉宁, 张家明, 杨宏飞, 等. 基于语义分割网络的AGV路径规划算法[J]. 电子测量与仪器学报, 2023, 37(07): 121-130.

李峥. 基于ROS平台的导航机器人路径规划研究[J]. 现代信息科技, 2020, 4(14): 163-165+168.

陈晨. 改进A*算法的机器人最短路径规划研究[J]. 计算机与数字工程, 2023, 51(08): 1697-1701.

Yangqi O, Yuexin F, inglan Z, et al. Improved A* Path Planning Method Based on the Grid Map[J]. Sensors, 2022, 22(16): 6198-6198.

Bijun T, Kaoru H, Xiangdong W, et al. Path Planning Based on Improved Hybrid A* Algorithm[J]. jaciii, 2021, 25(1): 64-72.

Cui G, Yin Y, Xu Q, et al. Efficient Path Planning for Automated Valet Parking: Integrating Hybrid A* Search with Geometric Curves[J]. International Journal of Automotive Technology, 2024, 26(1): 1-11.

Jin W, Li J, Lv X, et al. An optimized path planning approach for automatic parking using hybrid A* bidirectional search[J]. Applied Intelligence, 2024, 55(2): 201-201.

白俊卿, 魏雪涛, 张红猛. 基于改进混合A*算法的自动泊车路径规划方法研究[J]. 计算机测量与控制, 2025, 33(1): 226-234.

田雨, 林松, 房殿军, 等. 基于改进Hybrid A*的非完整约束机器人路径规划算法研究[J]. 机械, 2023, 50(3): 63-71.

Lin X, Baoye S, Maoyong C. A new approach to optimal smooth path planning of mobile robots with continuous-curvature constraint[J]. Systems Science & Control Engineering, 2021, 9(1): 138-149.

刘尚昆. 改进A*算法的智能机器人路径规划研究[D]. 哈尔滨理工大学, 2022.

Xie D, Xu Y, Wang R. Obstacle detection and tracking method for autonomous vehicle based on three-dimensional LiDAR[J]. International Journal of Advanced Robotic Systems, 2019, 16(2): 1729881419831587-1729881419831587.

谢德胜, 徐友春, 王任栋, 等. 基于三维激光雷达的无人车障碍物检测与跟踪[J]. 汽车工程, 2018, 40(08): 952-959.

许明, 张娟, 方志军. 自适应道路模型的非结构化道路检测算法[J]. 传感器与微系统, 2020, 39(1): 132-135.

张长勇, 韩梁. 基于优化DBSCAN的激光雷达障碍物检测[J]. 激光与光电子学进展, 2022, 59(12): 516-524.

Yang W, Dong S, Li D. The Research of 3D Point Cloud Data Clustering Based on MEMS Lidar for Autonomous Driving[J]. International Journal of Automotive Technology, 2024, 25(5): 1251-1262.

Sunil K, Afzal S. An intelligent optimize path planner for efficient mobile robot path planning in a complex terrain [J]. Microsystem Technologies, 2022, 29(4): 469-487.

Gong X, Gao Y, Wang F, et al. A Local Path Planning Algorithm for Robots Based on Improved DWA[J]. Electronics, 2024, 13(15): 2965-2965.

蒋华扬, 吴涛, 任建鑫, 等. 基于改进DWA算法的移动机器人局部路径规划[J]. 机械设计与研究, 2024, 40(5): 229-234.

袁新亚, 戴娟,孙胜强,等. 融合A*和DWA算法的移动机器人路径规划方法[J]. 电子测量技术, 2024, 47(4): 95-103.

Wang Z, Li G. Research on Path Planning Algorithm of Driverless Ferry Vehicles Combining Improved A* and DWA[J]. Sensors, 2024, 24(13): 4041-4041.

朱洪波, 殷宏亮. 煤矿救援机器人路径规划研究[J]. 工矿自动化, 2024, 50(12): 145-154.

熊冬杪, 宁勇强, 王嘉炜, 等. 基于改进DWA的移动机器人局部路径规划[J]. 工业控制计算机, 2024, 37(08): 102-104.

董翼宁, 曹景胜, 李刚. 融合A-Star与DWA双优化算法的自动引导车路径规划[J]. 科学技术与工程, 2023, 23(30): 12994-13001.

An G W, Qiang J Z. Research on formation navigation of USV based on improved A* and DWA fusion algorithm [J]. Journal of Computational Methods in Sciences and Engineering, 2024, 24(4-5): 3191-3209.