基于语义物体尺寸链的改进自适应蒙特卡洛动态定位研究

doi:10.3901/JME.2024.15.049

机械工程学报 ›› 2024, Vol. 60 ›› Issue (15): 49-59.doi: 10.3901/JME.2024.15.049

扫码分享

基于语义物体尺寸链的改进自适应蒙特卡洛动态定位研究

蒋林^1,2, 李云飞¹, 汤勃^1,3, 刘奇¹, 郭宇飞³, 赵慧^2,3

1. 武汉科技大学冶金装备及其控制教育部重点实验室武汉 430081;
2. 武汉科技大学机器人与智能系统研究院武汉 430081;
3. 武汉科技大学机械传动与制造工程湖北省重点实验室武汉 430081

收稿日期:2023-08-25 修回日期:2023-12-23 出版日期:2024-08-05 发布日期:2024-09-24
作者简介:蒋林，男，1976年出生，博士，教授，博士研究生导师。主要研究方向为室内移动机器人地图构建、定位、导航。E-mail：jianglin76@wust.edu.cn
李云飞，男，2000年出生，博士研究生。主要研究方向为室内移动机器人语义地图构建及其定位。E-mail：liyunfei20180607@163.com
汤勃，男，1973年出生，博士，教授，硕士研究生导师。主要研究方向为机器视觉、机器学习、机构运动学和动力学等。E-mail：tang1017@163.com
刘奇，男，1996年出生，硕士研究生。主要研究方向为室内移动机器人物体识别、语义地图构建。E-mail：liuqi_xl@163.com
郭宇飞，男，1985年出生，博士，副教授，硕士研究生导师。主要研究方向为非线性鲁棒控制理论及其在机器人、自动化等领域的应用。E-mail：guoyufei@wust.edu.cn
赵慧(通信作者)，女，1973年出生，博士，教授，硕士研究生导师。主要研究方向为智能机器人、鲁棒控制及电液伺服系统理论与应用。E-mail：zhwust@163.com
基金资助:
国家重点研发计划(2019YFB1310000)、国家自然科学基金(51874217)和湖北省重点研发计划(2020BAB098)资助项目。

Dynamic Localization Research on Improved AMCL Based on the Dimensional Chain of Semantic Objects

JIANG Lin^1,2, LI Yunfei¹, TANG Bo^1,3, LIU Qi¹, GUO Yufei³, ZHAO Hui^2,3

1. Key Laboratory of Metallurgical Equipment and Control Technology, Ministry of Education, Wuhan University of Science and Technology, Wuhan 430081;
2. Institute of Robotics and Intelligent Systems, Wuhan University of Science and Technology, Wuhan 430081;
3. Hubei Key Laboratory of Mechanical Transmission and Manufacturing Engineering, Wuhan University of Science and Technology, Wuhan 430081

Received:2023-08-25 Revised:2023-12-23 Online:2024-08-05 Published:2024-09-24

摘要/Abstract

摘要： 为解决原始自适应蒙特卡洛定位(Adaptive Monte Carlo localization,AMCL)在相似动态环境下极易重定位失败的问题，首先融合机器人里程计信息、激光信息以及视觉信息，基于Gmapping算法完成机器人的即时定位与栅格地图的创建，其次将KinectV2彩色图像输入到目标检测识别方法中获取图像中物体检测框与类别，并结合GrabCut算法实现语义分割，利用配准数据计算出彩色图像中的物体在深度图像中对应位置，然后将获取到的KinectV2深度数据转换成点云信息并投影转换至全局地图坐标系下，得到物体语义图，并充分发掘各语义物体之间的关系，构建语义物体尺寸链，并提出一种快速有效的尺寸链检索方法，同时采用贝叶斯方法来减小误检测和重复检测对物体语义图的影响，将物体语义图和栅格地图进行原点重合以及位姿对齐，从而构建二维语义栅格地图，提出一种基于语义物体尺寸链的改进AMCL重定位算法，最后通过大量真实环境下的重定位实验验证了所提方法在相似动态环境中的优越性能。

关键词: 相似动态环境, 语义物体尺寸链, 贝叶斯方法, 二维语义栅格地图, 重定位

Abstract: The aim is to solve the problem that the original adaptive Monte Carlo localization is prone to re-localization failure in similar dynamic environment. Firstly, the odometer, laser radar and vision information are fused, and the instant localization of the robot and the creation of grid map are completed based on the Gmapping algorithm. Secondly, the KinectV2 color image is input into the object detection and recognition method to obtain the boxes and categories of object detection in the image, and the semantic segmentation is realized by combining GrabCut algorithm. The corresponding position of the object in the depth image is calculated by using the registration data. Then, the acquired KinectV2 depth data is converted into point cloud information and projected onto the global map coordinate system to obtain the object semantic map. The relationship between the semantic objects is fully explored, and the dimension chain of semantic objects is constructed. And a fast and effective retrieval method of dimension chain is proposed. At the same time, the Bayesian method is used to reduce the influence of false detection and repeated detection on object semantic map. The semantic map of object and grid map are aligned by origin coincidence and pose alignment, so as to construct two-dimensional semantic grid map. The research proposes an improved AMCL re-localization algorithm base on the dimension chain of semantic objects. Finally, a large number of re-localization experiments in real environments verify the superior performance of the method in similar and dynamic environments.

中图分类号:

TP242

蒋林, 李云飞, 汤勃, 刘奇, 郭宇飞, 赵慧. 基于语义物体尺寸链的改进自适应蒙特卡洛动态定位研究[J]. 机械工程学报, 2024, 60(15): 49-59.

JIANG Lin, LI Yunfei, TANG Bo, LIU Qi, GUO Yufei, ZHAO Hui. Dynamic Localization Research on Improved AMCL Based on the Dimensional Chain of Semantic Objects[J]. Journal of Mechanical Engineering, 2024, 60(15): 49-59.

参考文献

[1] 赵洋.基于深度学习的语义同步定位与地图构建[D].济南:山东大学，2018. ZHAO Yang.Semantic synchronization localization and map construction based on deep learning[D].Jinan:Shandong University，2018.
[2] 白云汉.基于SLAM算法和深度神经网络的语义地图构建研究[J].计算机应用与软件，2018，35(1):183-190. BAI Yunhan.Research on semantic map construction based on SLAM algorithm and deep neural network[J].Computer Applications and Software，2018，35(1):183-190.
[3] 邹斌，林思阳，尹智帅.基于YOLOv3和视觉SLAM的语义地图构建[J].激光与光电子学进展，2020，57(20):124-130. ZOU Bin，LIN Siyang，YIN Zhishui.Semantic map construction based on YOLOv3 and visual SLAM[J].Laser & Optoelectronics Progress，2020，57(20):124-130.
[4] SONG S，YU F，ZENG A，et al.Semantic scene completion from a single depth image[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Honolulu，HI，USA:IEEE，2017:190-198.
[5] LI X，BELAROUSS R.Semi-dense 3D semantic mapping from monocular slam[J].arXiv Preprint arXiv，2016:1611.04144.
[6] KUNDU A，YIN L，DELLAERT F，et al.Joint semantic segmentation and 3D reconstruction from monocular video[C]//European Conference on Computer Vision (ECCV).Zurich，Switzerland:Springer，Cham，2014:703-718.
[7] MCCORMAC J，HANDA A，DAVISON A，et al.Semanticfusion:Dense 3D semantic mapping with convolutional neural networks[C]//2017 IEEE International Conference on Robotics and Automation (ICRA).Singapore:IEEE，2017:4628-4635.
[8] LIU Z，VON WICHERT G.Extracting semantic indoor maps from occupancy grids[J].Robotics and Autonomous Systems，2014，62(5):663-674.
[9] 蒋林，刘奇，雷斌，等.激光与视觉融合识别并构建语义地图改善定位性能[J].中国激光，2022，49(19):3-10. JIANG Lin，LIU Qi，LEI Bin，et al.Laser and vision fusion recognition and semantic map construction to improve localization performance[J].Chinese Journal of Lasers，2022，49(19):3-10.
[10] 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):306-314. JIANG Lin，XIANG Chao，ZHU Jianyang，et al.Particle filter relocation with semantic likelihood estimation[J].Acta Electronica Sinica，2021，49(2):306-314.
[11] ZHANG L，ZAPATA R，LÉPINAY P.Self-adaptive Monte Carlo localization for mobile robots using range finders[J].Robotica，2012，30(2):229-244.
[12] CHOI J，MAURER M.Simultaneous localization and mapping based on the local volumetric hybrid map[C]//2015 IEEE Intelligent Vehicles Symposium.Seoul，Korea (South):IEEE，2015:561-566.
[13] 杨记周.一种室内移动机器人定位和路径规划的算法优化[D].合肥:中国科学技术大学，2019. YANG Jizhou.An algorithm optimization for indoor mobile robot positioning and path planning[D].Hefei:University of Science and Technology of China，2019.
[14] 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):5-8. HU 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):5-8.
[15] 胡钊政，刘佳蕙，黄刚，等.融合WiFi、激光雷达与地图的机器人室内定位[J].电子与信息学报，2021，43(8):2308-2316. HU Zhaozheng，LIU Jiahui，HUANG Gang，et al.Robot indoor localization based on WiFi，lidar and map[J].Journal of Electronics and Information Technology，2021，43(8):2308-2316.
[16] YU S，YAN F，ZHUANG Y，et al.A deep-learning-based strategy for kidnapped robot problem in similar indoor environment[J].Journal of Intelligent & Robotic Systems，2020，100(4):765-775.
[17] 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. FANG 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.
[18] ZHAO X，ZUO T，HU X.OFM-SLAM:A visual semantic SLAM for dynamic indoor environments[J].Mathematical Problems in Engineering，2021，2021(1):1-16.
[19] 申文杰，任永杰，朱博源，等.结合人工信标和视觉SLAM的自主快速定位方法[J].机械工程学报，2022，58(14):137-145. SHEN Wenjie，REN Yongjie，ZHU Boyuan，et al.Autonomous fast localization method combined with manual beacon and visual SLAM[J].Journal of Mechanical Engineering，2022，58(14):137-145.
[20] 杨明辉，吴垚，张勇，等.室内动态环境下的移动机器人自主避障策略[J].中南大学学报，2019，50(8):1833-1839. YANG Minghui，WU Yao，ZHANG Yong，et al.Autonomous obstacle avoidance strategy of mobile robot in indoor dynamic environment[J].Journal of Central South University，2019，50(8):1833-1839.
[21] 邵延华，张铎，楚红雨，等.基于深度学习的YOLO目标检测综述[J].电子与信息学报，2022，44(10):3697-3708. SHAO Yanhua，ZHANG Duo，CHU Hongyu，et al.Review of YOLO object detection based on deep learning[J].Journal of Electronics and Information Technology，2022，44(10):3697-3708.

基于语义物体尺寸链的改进自适应蒙特卡洛动态定位研究

Dynamic Localization Research on Improved AMCL Based on the Dimensional Chain of Semantic Objects

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 5

编辑推荐

Metrics

本文评价

[1]	杨乐昌, 韩东旭, 王丕东. 基于Wasserstein距离测度的非精确概率模型修正方法[J]. 机械工程学报, 2022, 58(24): 300-311.
[2]	白森洋, 程志君, 郭波, 杨勇. 基于实时可靠度评估的航空产品序贯检测模型[J]. 机械工程学报, 2019, 55(2): 177-185.
[3]	姜潮, 张旺, 韩旭. 基于Copula函数的证据理论相关性分析模型及结构可靠性计算方法[J]. 机械工程学报, 2017, 53(16): 199-209.
[4]	彭卫文;黄洪钟;李彦锋;杨园鉴;李海庆. 基于数据融合的加工中心功能铣头贝叶斯可靠性评估[J]. , 2014, 50(6): 185-191.
[5]	凌光;戴怡;王仲民. 面向数控系统可靠性评估的最大熵先验信息解[J]. , 2012, 48(6): 157-161.