Improved AMCL Relocation Algorithm based on Semantic Map with Corner Information
-
摘要: 针对原始自适应蒙特卡洛定位(Adaptive monte carlo localization，AMCL)算法仅利用激光信息存在的缺陷，提出一种基于激光与视觉融合的语义地图进行全局定位，该语义地图融合基于深度学习的目标检测方法提取环境中的墙角语义；利用建立的包含墙角信息的二维语义栅格地图，结合视觉预定位方法及角点周围语义信息表来提高算法全局初始定位的效率和准确性，使得移动机器人可以在少量先验信息和运动的情况下更迅速地实现定位。提出视觉预定位的方法，改进了粒子权重更新方式，再同步结合AMCL算法与环境地图匹配进行精定位。最后通过搭建的移动机器人在不同场景下进行对比试验，验证了该方法的有效性。
针对原始自适应蒙特卡洛定位(adaptive monte carlo localization，AMCL)算法仅利用激光信息存在的缺陷，提出一种基于激光与视觉融合的语义地图进行全局定位，该语义地图融合基于深度学习的目标检测方法提取环境中的墙角语义；利用建立的包含墙角信息的二维语义栅格地图，结合视觉预定位方法及角点周围语义信息表来提高算法全局初始定位的效率和准确性，使得移动机器人可以在少量先验信息和运动的情况下更迅速地实现定位。提出视觉预定位的方法，改进了粒子权重更新方式，再同步结合AMCL算法与环境地图匹配进行精定位。最后通过搭建的移动机器人在不同场景下进行对比试验，验证了该方法的有效性。
Abstract: Aiming at the defects of the original AMCL positioning algorithm using only laser information, a semantic map based on laser and vision fusion is proposed for global positioning. The semantic map is fused with a target detection method based on deep learning to extract the corner semantics in the environment; use The established two-dimensional semantic grid map containing corner information, combined with the visual pre-positioning method and the semantic information table around the corner points to improve the efficiency and accuracy of the algorithm's global initial positioning, so that the mobile robot can be used in the situation of a small amount of prior information and motion To achieve positioning more quickly. The method of visual pre-positioning is proposed, the particle weight update method is improved, and the AMCL algorithm is combined with the environment map to perform precise positioning. Finally, a comparative experiment of the built mobile robot in different scenarios verifies the effectiveness of the method.-
Key words:
- corner information /
- semantic information table /
- global pre-positioning /
- improved AMCL /
- semantic map
-
表 1 角点周围语义查找表
角点 椅子 垃圾桶 门 柜子 d1/m d2/m d3/m d4/m ① 0 1 1 2 0 0.42 1.10 1.44
1.57② 0 1 0 0 0 1.14 0 0 ③ 1 0 1 2 2.8 0 0.92 1.46
1.63
下载: 导出CSV
-
[1] 邢志伟, 朱晓蕊, 何超. 基于行人行为学习的机器人同时定位与可通行区域制图[J]. 机械工程学报, 2019, 55(11): 36-45. doi: 10.3901/JME.2019.11.036XING Zhiwei, ZHU Xiaorui, HE Chao. Simultaneous localization and traversable region mapping based on pedestrian behavior learning[J]. Journal of Mechanical Engineering, 2019, 55(11): 36-45. doi: 10.3901/JME.2019.11.036 [2] 罗元, 庞冬雪, 张毅, 等. 基于自适应多提议分布粒子滤波的蒙特卡洛定位算法[J]. 计算机应用, 2016, 36(8): 2352-2356. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY201608055.htmLUO Yuan, PANG Dongxue, ZHANG Yi, et al. Monte carlo localization algorithm based on particle filter with adaptive multi-proposal distribution[J]. Journal of Computer Applications, 2016, 36(8): 2352-2356. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY201608055.htm [3] DELLAERT F, FOX D, BURGARD W, THRUN S. Monte Carlo localization for mobile robots[C]// Proceedings of the 1999 IEEE International Conference on Robotics and Automation, May 10-15, 1999, Detroit, USA, 1999 : 1322-1328. [4] ZHANG L, ZAPATA R, LEPINAY P. Self-adaptive monte carlo localization for mobile robots using range finders[J]. Robotica, 2012, 30(2): 229-244. doi: 10.1017/S0263574711000567 [5] HANTEN R, BUCK S, OTTE S, et al. Vector-AMCL: Vector based adaptive monte carlo localization for indoor maps[C]//14th International Conference on Intelligent Autonomous Systems, July 3-7, 2016. Shanghai, China, 2016: 403-416. [6] CHOI J, MAURER M. Hybrid map-based SLAM with rao-black wellized particle filters[C]// 17th International Conference on Information Fusion, July 7-10, 2014, Salamanca, Spain, 2014 : 1-6. [7] PARK S, ROH K S. Coarse-to-fine localization for a mobile robot based on place learning with a 2D range scan[J]. IEEE Transactions on Robotics, 2016(3): 1-17. [8] 胡章芳, 曾林全, 罗元, 等. 融入二维码信息的自适应蒙特卡洛定位算法[J]. 计算机应用, 2019, 39(4): 989-993. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY201904010.htmHU Zhangfan, ZENG Linquan, LUO Yuan, et al. Adaptive monte-carlo localization algorithm integrated with two-dimensional code information[J]. Journal of Computer Applications, 2019, 39(4): 989-993. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY201904010.htm [9] 郑文磊, 程磊, 余秋月, 等. 基于自适应观测模型的移动机器人室内蒙特卡洛动态定位系统研究[J]. 高技术通讯, 2017, 27(Z1): 848-855. https://www.cnki.com.cn/Article/CJFDTOTAL-GJSX2017Z1010.htmZHENG Wenlei, CHENG Lei, YU Qiuyue, et al. Research on an indoor monte carlo dynamic positioning system for mobile robots based on adaptive observation model[J]. High Technology Letters, 2017, 27(Z1): 848-855. https://www.cnki.com.cn/Article/CJFDTOTAL-GJSX2017Z1010.htm [10] 刘洞波, 刘国荣, 喻妙华, 等. 一种基于图像检索的机器人自定位方法[J]. 传感技术学报, 2010, 23(4): 548-552. https://www.cnki.com.cn/Article/CJFDTOTAL-CGJS201004022.htmLIU Dongbo, LIU Guorong, YU Miaohua, et al. A robot self-localization method based oil image retrieval system[J]. Chinese Journal of Sensors and Actuators, 2010, 23(4): 548-552. https://www.cnki.com.cn/Article/CJFDTOTAL-CGJS201004022.htm [11] 贾云辉. 基于ROS系统的移动机器人室内定位方法研究[D]. 天津: 天津理工大学, 2019.JIA Yunhui. Research on indoor location method of mobile robot based on ROS system[D]. Tianjin: Tianjin University of Technology, 2019. [12] 肖亚龙, 张士庚, 王建新. 一种基于多维标度和区域细化的无线室内定位方法[J]. 计算机学报, 2017, 40(8): 1918-1932. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJX201708013.htmXIAO Yalong, ZHANG Shigeng, WANG Jianxin. An indoor localization algorithm based on multidi-mensional scaling and region refinement[J]. Chinese Journal of Computers, 2017, 40(8): 1918-1932. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJX201708013.htm [13] XIAO Linhui, WANG Jinge, QIU Xiaosong, et al. Dynamic-SLAM: Semantic monocular visual localization and mapping based on deep learning in dynamic environment[J]. Robotics and Autonomous Systems, 2019, 117: 1-16. [14] 房立金, 刘博, 万应才. 基于深度学习的动态场景语义SLAM[J]. 华中科技大学学报, 2020, 48(1): 121-126. https://www.cnki.com.cn/Article/CJFDTOTAL-HZLG202001022.htmFANG Lijin, LIU Bo, WAN Yingcai. Semantic SLAM based on deep learning in dynamic scenes[J]. Journal of Huazhong University of Science and Technology, 2020, 48(1): 121-126. https://www.cnki.com.cn/Article/CJFDTOTAL-HZLG202001022.htm [15] 席志红, 韩双全, 王洪旭. 基于语义分割的室内动态场景同步定位与语义建图[J]. 计算机应用, 2019, 39(10): 2847-2851. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY201910010.htmXI Zhihong, HAN Shuangquan, WANG Hongxu. Simultaneous localization and semantic mapping of indoor dynamic scene based on semantic segmentation[J]. Journal of Computer Applications, 2019, 39(10): 2847-2851. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY201910010.htm [16] 张威. 基于物体语义信息的室内视觉SLAM研究[D]. 济南: 山东大学, 2019.ZHANG Wei. Indoor visual SLAM based on object semantic information[D]. Jinan: Shandong University, 2019. [17] CAI Lecai, YE Yuling, GAO Xiang, et al. An improved visual SLAM based on affine transformation for ORB feature extraction[J]. Optik - International Journal for Light and Electron Optics, 2021, 227: 1-15. [18] 杨爽, 曾碧, 何炜婷. 融合语义激光与地标信息的SLAM技术研究[J]. 计算机工程与应用, 2020, 56(18): 262-271. https://www.cnki.com.cn/Article/CJFDTOTAL-JSGG202018038.htmYANG Shuang, ZENG Bi, HE Weiting. Research on SLAM technology combining semantic laser and landmark information[J]. Computer Engineering and Applications, 2020, 56(18): 262-271. https://www.cnki.com.cn/Article/CJFDTOTAL-JSGG202018038.htm [19] 魏乐麒. 基于环境语义信息的SLAM算法研究与实现[D]. 西安: 西安电子科技大学, 2018.WEI Leqi. A study and implementation of SLAM system algorithm based on semantic messages[D]. Xi'an: Xidian University, 2018. [20] ZHANG Chen, WANG Mei, YU Yunlei, et al. SLAM system based on tightly coupled visual-inertial[J]. Journal of Physics Conference Series, 2020, 1576(1): 1-7. [21] 蒋林, 向超, 朱建阳, 等. 加载语义似然估计的粒子滤波重定位[J]. 电子学报, 2021, 49(2): 315-323. https://www.cnki.com.cn/Article/CJFDTOTAL-DZXU202102013.htmJIANG Lin, XIANG Chao, ZHU Jianyang, et al. Particle filter relocation with semantic likelihood estimation[J]. Acta Electronica Sinica, 2021, 49(2): 315-323. https://www.cnki.com.cn/Article/CJFDTOTAL-DZXU202102013.htm -
下载:
点击查看大图
图(23) / 表(1)
计量
- 文章访问数: 49
- HTML全文浏览量: 13
- PDF下载量: 0
- 被引次数: 0
京公网安备 11010502036328号 京ICP备17033152号