定位功能与路径规划是AGV实现无人自动导航的关键技术。现如今由于AGV的广泛应用，“机器人绑架”或“迷失”事件时有发生，且当前单一路径规划算法也难以应对日益复杂的环境。针对以上两点问题，本文提出一种基于激光SLAM与OpenCV模板匹配的全局重定位算法，确保机器人在失去位姿信息后能够安全高效的进行重定位操作，并提出基于启发式搜索策略的改进蚁群算法进行路径规划，主要研究内容如下：

搭建AGV小车，使用STM32系列单片机作为下层控制单元，内置IMU传感器。AGV搭载激光雷达进行SLAM建图，并进行IMU数据与激光雷达数据的同步处理。采用栅格法作为本文的环境建模方法，为后续章节提供基础。

针对AGV定位算法，设计一种利用激光雷达SLAM建图与OpenCV模板匹配法的全局重定位算法，利用IMU处理激光SLAM中点云运动畸变，构建AGV周边环境地图(submap)，将其与全局地图进行匹配，利用OpenCV算法获取submap在全局地图中的信息位置，通过坐标变换关系反解出AGV在全局地图中的实际位姿信息。

针对AGV路径规划算法，本文主要针对蚁群算法进行改进。通过事先填充策略解决蚁群陷入“U型陷阱点”问题；采用启发式搜索策略解决蚁群算法在路径规划前期无需搜索问题，提高算法效率；改进传统蚁群算法的信息素更新策略，避免算法陷入局部最优解；最终采用贪心策略对算法整体进行冗余节点剔除，保证路径的平滑度。

搭建实际实验环境，并利用搭建好的全向麦克纳姆轮AGV实验平台进行实地实验测试，验证其算法的可行性。

Positioning function and path planning are the key technologies for AGV to achieve unmanned automatic navigation. Nowadays, due to the wide application of AGV, "robot kidnapping" or "lost" events occur from time to time, and the current single path planning algorithm is difficult to cope with the increasingly complex environment. In order to solve the above two problems, this paper proposes a global relocation algorithm based on laser SLAM and OpenCV template matching to ensure that the robot can safely and efficiently relocate after losing the pose information, and proposes an improved ant colony algorithm based on heuristic search strategy for path planning, the main research contents are as follows:

Build an AGV trolley, use STM32 series microcontroller as the lower control unit, and build a built-in IMU sensor. AGVs are equipped with LiDAR for SLAM mapping, and synchronous processing of IMU data and LiDAR data. The raster method is used as the environmental modeling method in this paper, which provides a basis for the subsequent chapters.

For the AGV positioning algorithm, a global relocation algorithm using LiDAR SLAM mapping and OpenCV template matching method was designed, the IMU was used to process the point cloud motion distortion in the laser SLAM, the AGV surrounding environment map (submap) was constructed, and it was matched with the global map, the OpenCV algorithm was used to obtain the information location of the submap in the global map, and the actual pose information of AGV in the global map was inversely solved through the coordinate transformation relationship.

For the AGV path planning algorithm, this paper mainly improves the ant colony algorithm. The problem of ant colony falling into "U-shaped trap point" is solved by pre-filling strategy, the heuristic search strategy is used to solve the problem that ant colony algorithm does not need to search in the early stage of path planning, and the algorithm

efficiency is improved, the pheromone update strategy of traditional ant colony algorithm is improved to avoid the algorithm falling into local optimal solution, and finally the greedy strategy is used to eliminate redundant nodes of the algorithm as a whole to ensure the smoothness of the path.

The actual experimental environment was built, and the field experiment was carried out by using the built omnidirectional Mecanum wheel AGV experimental platform to verify the feasibility of the algorithm.

[1]杨森.基于多传感器融合的AGV定位技术研究[D];浙江师范大学,2022.

[2]武启平,金亚萍,任平,等.自动导引车(AGV)关键技术现状及其发展趋势[J].2013,35(10):5.

[3]Cheeseman P, Smith R, Self M. A stochastic map for uncertain spatial relationship[C]//4th International Symposium on Robotic Research. 1987: 467-474.

[4]刘浩敏，章国锋，鲍虎军，基于单目视觉的同时定位与地图构建方法综述[J],计算机辅助设计与图形学学报，2016，28(6)：855-868.

[5]林辉灿，吕强，王国胜，等.基于VSLAM的自主移动机器人三维同时定位与地图构建[J].计算机应用，2017，37(10)：2884-2887.

[6]Liu Y, Xiong R, Wang Y, et al. Stereo Visual-inertial odometry with multiple Kalman filters ensemble[J]. IEEE Transactions on Industrial Electronics, 2016,63(10):6205-6216.

[7]Burri M, Oleynikova H, Achtelik M W,et al. Real-time visual-inertial mapping, re-localization and planning onboard mavs in unknown environments[C]//2015 IEEE/RSJ International Conference on Intelligent Robots and Systems(IROS). IEEE, 2015: 1872-1878.

[8]Lategahn H, Geiger A, Kitt B. Visual SLAM for autonomous ground vehicles[C]//2011 IEEE International Conference on Robotics and Automation. IEEE, 2011:1732-1737

[9]Jung J, Choi S, Choi H T,et al.Localization of AUVs using depth information of underwater structures from a monocular camera[C]//2016 13th International Conference on Ubiqutious Robots and Ambient Intelligence (URAI). IEEE,2016:444-446.

[10]Piao J C, KIM S D. Adaptive monocular visual-inertial SLAM for real-time augmented reality applications in mobile devices[J]. Sensors,2017,17(11):2567-2577.

[11]李然,王成,苏国中,等.星载激光雷达的发展与应用[J]科技导报.2007(14):3-5.

[12]罗骞，金琦珺. 智能汽车激光雷达感知技术现状与发展分析，科技与创新2020,（17）.

[13]刘莹莹，杨砚儒.激光雷达及其应用[J]. 高等职业教育(天津职业职业大学学报),2007(2):73-75

[14]张振涛,王伟,苏贵波.激光雷达的现状与发展趋势[J],科技信息.2012,(10):1.

[15]景慧. 全向移动机器人复杂场景重定位与跟踪控制研究及ROS实现[D],东北林业大学2021.

[16]Smith R C，Cheeseman P. On the representation and estimation of spatial uncertainty[J]. International Journal of Robotics Research,1986,5(4):56-68.

[17]Dissanayake G, Huang S, Wang Z, et al. A revire of recent developments in Simultaneous Localization and Mapping [C];6th International Conference on Industrial and Information Systems, 2011:477-482.

[18]Kumerle R, Grisetti G, Strasdat H, et al. G2o: A general framework for graph optimization [C]; IEEE International Conference on Robotics and Automation, 2011 : 3607-3613.

[19]Polok L, Solony M, Smrz P,et al. Incremental cholesky factorization for least squares problems in robotics [C]; IFAC Proceedings Volumes(IFAC-PaperOnline),2013:172-178.

[20]Triggs B, McLauchlan PF, Hartley R I, et al. Bundle adjustment-a modern synthesis [C]; Lecture Notes in Computer Science, 2000:298-372.

[21]Zhao Mingle, Zhou Dingfu, Song Xibin,Chen Xiuwan, Zhang Liangjun. DiT-SLAM:Real-time Dense Visual-inertial SLAM with implicit depth representation and Tightly-coupled graph optimization[J]. Sensors,2022,22(9):54-62.

[22]Czapla T , Wrona J. Technology Development of Military Applications of Unmanned Ground Vehicles[M]. Springer International Publishing,2013:293-309.

[23]Liu D, Duan J, Shi H. Astrong tracking square root central difference FastSLAM for unmanned intelligent vehicle with adaptive partial systematic resampling[J]. IEEE Transactions on Intelligent Transportation Systems,2016,17(11):3110-3120.

[24]Cho H, Kim E K, Kim S. Indoor SLAM application using geometric and ICP matching methods based on line features[J]. Robotics and Autonomous Systems, 2018, 100:206-224.

[25]刘欣，陈武. 基于视觉分析的机器人最优避障路径识别方法[J]. 西南大学学报2017,39(2):128-134.

[26]Li Y, Hu Z, Hu Y, et al. Integration of vision and topological self-localization for intelligent vehicles[J]. Mechatronics,2018,51:46-58.

[27]Yang T, Li P, Zhang H,et al. Monocular vision SLAM-based UAV autonomous landing in emergencies and unknown environments[J]. Electronics,2018,7(5):73.

[28]杨阔. 动态场景下的 2D SLAM方法研究[D]. 北京: 北京交通大学，2018.

[29]章大勇. 激光雷达/惯性组合导航系统的一致性于最优估计问题研究[D]. 国防科学技术大学，2010.

[30]王平,鲁可,曹云峰,等.微小型飞行器室内激光/INS融合定位方法研究[J].飞行器测控学报,2013,32(004):340-344.

[31]胡晓强，刘汉忠，贾良冠等.基于ROS平台的激光雷达Slam-GMapping的重定位方法研究[J].现代信息科技,2019,3(12)：159-161.

[32]Jun Wang，Yasutake Takahashi. SLAM Method Based on Independent Particle Filters for Landmark Mapping and Localization for Mobile Robot Based on HF-band RFID System[J]. Journal of Intelligent & Robotic Systems,2018,92(3):413-433.

[33]Fox D. Monte Carlo localization: efficient estimation for mobile robots[C]// Burgard W, Dellaert F, et al. National Conference on Artificial Intelligence,1999.

[34]Ronghua Luo, Huaqing Min, Maohai Li, et al. A Method for Active Global Localization in Multi-robot System[J]. International Journal of Advanced Robotic Systems,2008,5(3):269-278.

[35]D Fox, W Burgrad, S Thrun. Markov localization for mobile robots in dynamic environments[J]. Journal of Artificial Intelligence Research,1999,11:391-427.

[36]S Thrun, D Fos, W Burhard. Robot Monte Carlo localization for mobile robots[J]. Artificial Intelligence,2001,128(1):99-141.

[37]丁豪.移动机器人重定位方法研究及巡检系统开发[D].扬州大学,2021.

[38]李忠发，杨光，马磊等. 变电站巡检机器人重定位研究[J]. 计算机科学,2020,47(S1)：599-602.

[39]Hess W. Real-Time Loop Closure in 2D LIDAR SLAM [C]// Kohler D, Rapp H, et al. 2016 IEEE International Conference on Robotics and Automation (ICRA),2016.

[40]焦传佳，江明，徐劲松等. 基于激光信息的移动机器人定位研究[J].电子测量与仪器学报,2021,35(09):4-7.

[41]向超，蒋林，雷斌等. 基于环境语义信息的移动机器人重定位增强[J].武汉科技大学学报,2020，43(3)：224-229.

[42]M Raul, J Tardos. ORB-SLAM2: An open-source slam system for monocular, stereo, and RGB-D cameras[J]. IEEE Transactions on Robotics,2017,33(5):1225-1262.

[43]高翔. 视觉SLAM14讲从理论到实践[M] //张涛等. 北京：电子工业出版社，2019.

[44]N Radwan,A Valada, W Burgard. VLocNet++: Deep Mutitask Learning for Semantic Visual Localization and Odometry[J]. IEEE Robotics and Automation Letters,2018,3(4):4407-4414.

[45]A Valada. Deep Auxiliary Learning for Visual Localization and Odometry [C]// N Radwan, W Burgard. Robotics and Automation, IEEE International Conference on. IEEE,2018.

[46]林韩熙，向丹，欧阳剑，兰晓东. 移动机器人路径规划算法的研究综述[J]. 计算机工程与应用，2021,57(18):2-9.

[47]Krzysztof Nowak.Application of a nonlocal grid model for analysis of the creep damage of metals[J].International Journal of Damage Mechanics,2020,29(05):780-797.

[48]白俊涛.井下救援机器人感知系统设计与实现[D].西安科技大学,2020.

[49]牛秦玉,李美凡,赵勇.改进人工势场法的AGV路径规划算法研究[J].机械制造与自动化,2022,67(13):19-24.

[50]Liu Y，Li J，et al. Robot Path Planning Based on Dijkstra’s Algorithm and Genetic Algorithm[C]//Lecture Notes in Electrical Engineering,2018,422:397-408.

[51]Khatib. Real-time obstacle avoidance for manipulators and mobile robots[J].1985,2:500-505.

[52]李晓旭,马兴录,王先鹏.移动机器人路径规划算法综述[J].计算机测量与控制,2022,30(07):4-9.

[53]Sangeetha V, Ravichandran K S, Shekhar S, et al. An Intelligent Gain based Ant Colony Optimisation Method for Path Planning of Unmanned Ground Vehicles[J]. Defence science journal, 2019,69(2):167-172.

[54]Lumelsky V, Stepanov A. Dynamic path planning for a mobile automaton with limited information on the environment[J]. IEEE Transaction on Automatic Control,1986,31(11):1058-1063.

[55]Colorni A, Dorigo M, Maniezzo V. Distributed Optimization by Ant Colonies[C] // Proceedings of ECAL91-European Conference on Artificial Life. 1991.

[56]吉泽勇.井下移动机器人目标识别与自主避障研究[D].中国矿业大学,2022.

[57]陈卓.自动导引小车(AGV)运动学模型和动力学仿真[J].河南科技,2015,555(01):47-49.

[58]蒙松.全向AGV的位姿估计与定位研究[D].西安科技大学,2021.

[59]王子轩.动态场景下视觉惯性导航和语义地图构建[D].天津大学,2020.

[60]李辰旭.基于激光SLAM的AGV建图与定位技术研究[D].武汉纺织大学,2023.

[61]张保,张安思,梁国强,张旭东,王从宝.激光雷达室内定位技术研究及应用综述[J].激光杂志,2023,44(03):4-6.

[62]孙仲旭.基于激光雷达的AGV定位技术研究[D].哈尔滨工业大学,2020.

[63]Remote sensing;Study results from Wuhan University Update Understanding of remote sensing (Improved IMU Preintegration with gravity change and earth rotation for Optimization-based GNSS/VINS)[J]. Journal of enginerring,2020,45(26):45-56.

[64]张杰,王璐.基于二维激光雷达的AGV小车定位研究[J].佳木斯大学学报(自然科学版),2019,37(06):947-950.

[65]冯振.基于前后相机视觉信息融合的移动机器人局部定位算法[D].哈尔滨工业大学,2019.