检索结果-内蒙古大学图书馆

MoTI: A Multi-Stage algorithm for Moving Object Identification in slam

SENSORS 2023年第18期23卷 7911-7911页

作者： Hu, Changqing Liu, Manlu Zhang, Su Xie, Yu Tan, Liguo Southwest Univ Sci & Technol Sch Informat Engn Mianyang 621010 Peoples R China Southwest Univ Sci & Technol Robot Technol Used Special Environm Key Lab Sichua Mianyang 621010 Peoples R China Cent South Univ Sch Traff Transportat Engn Changsha 410000 Peoples R China Harbin Inst Technol Lab Space Environm & Phys Sci Harbin 150001 Peoples R China

Simultaneous localization and mapping (slam) algorithms are widely applied in fields such as autonomous driving and target tracking. However, the effect of moving objects on localization and mapping remains a challenge in natural dynamic scenarios. To overcome this challenge, this paper proposes an algorithm for dynamic point cloud detection that fuses laser and visual identification data, the multi-stage moving object identification algorithm (MoTI). The MoTI algorithm consists of two stages: rough processing and precise processing. In the rough processing stage, a statistical method is employed to preliminarily detect dynamic points based on the range image error of the point cloud. In the precise processing stage, the radius search strategy is used to statistically test the nearest neighbor points. Next, visual identification information and point cloud registration results are fused using a method of statistics and information weighting to construct a probability model for identifying whether a point cloud cluster originates from a moving object. The algorithm is integrated into the front-end of the LOAM system, which significantly improves the localization accuracy. The MoTI algorithm is evaluated on an actual indoor dynamic environment and several KITTI datasets, and the results demonstrate its ability to accurately detect dynamic targets in the background and improve the localization accuracy of the robot.

关键词： moving object detection multi-sensor fusion point cloud processing slam algorithm

来源：评论

学校读者我要写书评

暂无评论

FFD-slam: A Real-Time Visual slam Toward Dynamic Scenes with Semantic and Optical Flow Information

引用

JOURNAL OF ADVANCED COMPUTATIONAL INTELLIGENCE AND INTELLIGENT INFORMATICS 2024年第3期28卷 586-594页

作者： Zhang, Hao Wang, Yu Zhong, Tianjie Dong, Fangyan Chen, Kewei Ningbo Univ Fac Mech Engn & Mech 818 Fenghua Rd Ningbo 315211 Zhejiang Peoples R China China Acad Safety Sci & Technol Bldg A132 Beiyuan Rd Beijing 100012 Peoples R China

To solve the problem of poor localization accuracy and robustness of visual simultaneous localization and mapping (slam) systems in highly dynamic environments, this paper proposes a dynamic visual slam algorithm called FFD-slam that fuses the target detection network with the optical flow method. The algorithm considers ORB-slam2 as the basic framework, joins the semantic thread in parallel with its tracking thread, initially obtains the set of feature points through the real-time detection of dynamic objects in the environment through YOLOv5 in the semantic thread, then filters the set of feature points obtained in the semantic thread through the optical flow module, and finally utilizes the remaining static feature points for the matching calculation. Experiments showed that the proposed algorithm showed an improvement of approximately 97% in the localization accuracy compared with the ORB-slam2 algorithm in a highly dynamic environment, which effectively improves the localization accuracy and robustness of the system. The proposed algorithm also showed a higher real-time performance compared with some excellent dynamic slam algorithms.

关键词： dynamic object detection optical flow filter- ing image feature points elimination slam algorithm

来源：评论

学校读者我要写书评

暂无评论

Research on Autonomous Navigation System of Agricultural Quadruped Robot 21

Research on Autonomous Navigation System of Agricultural Qua...

引用

21st IEEE International Conference on Mechatronics and Automation (IEEE ICMA)

作者： Zhang, Jie Wang, Xinyue Zheng, Liang Jilin Agr Sci & Technol Univ Sch Elect & Informat Engn 77 Hanlin Rd Jilin 132101 Jilin Peoples R China

ISBN: (纸本)9798350388084;9798350388077

The slam is the key technology in the field of robotics, it can use lidar to scan the environment and extract feature points, realize the autonomous positioning and map building robot. In recent years, with the continuous development of robotics and lidar technology, laser slam has been widely studied and applied. This paper mainly from the quadruped robot path planning, environment perception, positioning and navigation and other key issues are deeply studied and discussed. First of all, the application field and basic principle of agricultural quadruped robot based on slam technology are analyzed and studied. Then, the main algorithm function and system implementation of agricultural quadruped robot based on slam technology are analyzed in detail. Finally, a more reasonable design algorithm of autonomous navigation system of quadruped robot is proposed, and the algorithm is applied to the robot for testing. After testing, the function of autonomous navigation system of quadruped robot meets the design requirements.

关键词： slam algorithm Agricultural quadruped robot Map building Autonomous navigation system

来源：评论

学校读者我要写书评

暂无评论

Advancements in underground mine surveys by using slam-enabled handheld laser scanners

引用

SURVEY REVIEW 2022年第385期54卷 363-374页

作者： Ellmann, Artu Kutimets, Kaia Varbla, Sander Vali, Erik Kanter, Sander Tallinn Univ Technol Sch Engn Dept Civil Engn & Architecture Ehitajate Rd 5 EE-19086 Tallinn Estonia Tallinn Univ Technol Sch Sci Dept Geol Ehitajate Td 5 EE-19086 Tallinn Estonia

Applicability of slam (simultaneous localization and mapping) technology for mine surveys and subsequent 3D modelling of post-extracted surfaces is assessed. The resulting surface geometry is validated via terrestrial laser scanner (TLS) acquired reference data. Typical discrepancies remained within 2 and 5 cm in horizontal and vertical directions, respectively. Discrepancies between TLS, slam-enabled handheld scanner and conventional surveying results are small and fully satisfy the contemporary accuracy requirements, yet evidence that the conventional mine survey results are affected by the subjectivity of the surveyors. The slam-enabled laser scanning hence appears to be the most suitable method for underground mining surveys.

关键词： Mobile laser scanning Handheld laser scanner slam algorithm Oil shale mine Estonia

来源：评论

学校读者我要写书评

暂无评论

Underground oil shale mine surveying using handheld mobile laser scanners

引用

OIL SHALE 2021年第1期38卷 42-64页

作者： Kutimets, Kaia Ellmann, Artu Vali, Erik Kanter, Sander Tallinn Univ Technol Sch Engn Dept Civil Engn & Architecture Ehitajate Tee 5 EE-19086 Tallinn Estonia Tallinn Univ Technol Dept Geol Sch Sci Ehitajate Tee 5 EE-19086 Tallinn Estonia

The applicability of a handheld mobile laser scanner (MLS) in oil shale mine surveys and subsequent three-dimensional modelling of postextracted surfaces is assessed. Recommendations for optimizing the acquisition and processing of MLS data and visualization of the results are given. The resulting surface geometry accuracy is validated via terrestrial laser scanner (TLS) reference data. Typical discrepancies between TLS and MLS data points remain within 2 and 5 cm in horizontal and vertical directions, respectively. The area of pillars and volumes of the extracted material are estimated by data analysis. The results are compared with those of the conventional mining survey. The detected discrepancies evidence that the laser scanning results provide a realistic outcome due to the evenly and densely spaced points within the point cloud. The result discrepancies between the tested surveying technologies are small and fully satisfy contemporary accuracy requirements. However, the handheld mobile laser scanning appears to be the most suitable method for underground mining surveys. The survey results enable reduction of mining losses and improvementt of the design of mining geometry.

关键词： mobile laser scanning handheld laser scanner slam algorithm mining losses underground mine Estonia

来源：评论

学校读者我要写书评

暂无评论

Object registration in semi-cluttered and partial-occluded scenes for augmented reality

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2019年第11期78卷 15079-15099页

作者： Gao, Qing Hong Wan, Tao Ruan Tang, Wen Chen, Long Xian Polytech Univ Sch Elect & Informat Xian Shaanxi Peoples R China Univ Bradford Fac Engn & Informat Bradford BD7 1DP W Yorkshire England Bournemouth Univ Fac Sci & Technol Poole BH12 5BB Dorset England

This paper proposes a stable and accurate object registration pipeline for markerless augmented reality applications. We present two novel algorithms for object recognition and matching to improve the registration accuracy from model to scene transformation via point cloud fusion. Whilst the first algorithm effectively deals with simple scenes with few object occlusions, the second algorithm handles cluttered scenes with partial occlusions for robust real-time object recognition and matching. The computational framework includes a locally supported Gaussian weight function to enable repeatable detection of 3D descriptors. We apply a bilateral filtering and outlier removal to preserve edges of point cloud and remove some interference points in order to increase matching accuracy. Extensive experiments have been carried to compare the proposed algorithms with four most used methods. Results show improved performance of the algorithms in terms of computational speed, camera tracking and object matching errors in semi-cluttered and partial-occluded scenes.

关键词： Augmented reality 3D object recognition and matching 3D point clouds slam algorithm

来源：评论

学校读者我要写书评

暂无评论

Semantic map construction based on monocular vision 11

Semantic map construction based on monocular vision

引用

11th International Conference on Intelligent Human-Machine Systems and Cybernetics (IHMSC)

作者： Xu, Yiming Dai, Qiuxia Gu, Haifeng Lu, Guan Gu, Juping Hua, Liang Nantong Univ Sch Elect Engn Nantong Jiangsu Peoples R China

ISBN: (纸本)9781728118598

The traditional environment maps built by robots include both metric ones and topological ones. These maps are navigation-oriented and not adequate for service robots to interact with or serve human users who normally rely on conceptual knowledge or semantic contents of the environment. Therefore, the construction of semantic maps becomes necessary for building an effective human-robot interface for service robots. In addition, due to the breakthrough results of the Deep Convolutional Neural Network in the field of target detection, this paper uses the DCNN to detect the target object effectively. Based on this, the slam algorithm is used to construct the map, and combined with the detection algorithm to realize the construction of the semantic map. Experiments show that the system can be applied to robot intelligent navigation.

关键词： semantic map target detection slam algorithm deep learning

来源：评论

学校读者我要写书评

暂无评论

A Fire Protection Robot System Based on slam Localization and Fire Source Identification 9

A Fire Protection Robot System Based on SLAM Localization an...

引用

IEEE 9th International Conference on Electronics Information and Emergency Communication (ICEIEC)

作者： Fan, Xiaojing Xin, Zhiqiang Li, Jiatong Li, Yifan Wan, Junhui Sun, Hao Yang, Zhixiong Yu, Muzhou Yang, Junfeng Cheng, Long Northeastern Univ Dept Comp & Commun Engn Qinhuangdao 066004 Hebei Peoples R China

ISBN: (纸本)9781728111902

When faced with various dangerous environments such as high temperature, lack of oxygen, and heavy smoke, fire-fighting robots can replace rescuers to enter the scene of the accident to extinguish fires and reduce casualties. This paper proposes a method combining Gmapping slam algorithm and fire source identification image processing algorithm to realize the functions of fire-fighting robot autonomous navigation and fire source identification detection. slam construction uses RBPF algorithm and effective particle number. The identification and location of the fire source point utilizes a series of image processing techniques. It can be seen from the simulation results that the map boundary established by the slam algorithm is obvious, and the terrain environment is well reproduced, and the positioning of the robot is realized. The fire source identification algorithm accurately identifies the area and contour of the flame, and realizes the intelligent detection of the fire source point. In addition, the fusion of the two algorithms enables the fire-fighting robot to independently detect the fire source in the indoor environment, effectively preventing the occurrence of fire accidents.

关键词： fire-fighting robot slam algorithm particle filter fire source identification reverse perspective

来源：评论

学校读者我要写书评

暂无评论

Implementation of an Embedded Testbed for Indoor slam 15

Implementation of an Embedded Testbed for Indoor SLAM

引用

15th IEEE/ACS International Conference on Computer Systems and Applications (AICCSA)

作者： Hanif, Muhammad Shehzad Bilal, Muhammad Munawar, Khalid Balamash, Abdullah Saeed King Abdulaziz Univ CEIES Jeddah 21589 Saudi Arabia King Abdulaziz Univ Dept Elect & Comp Engn Jeddah 21589 Saudi Arabia

ISBN: (纸本)9781538691205

Simultaneous Localization and Mapping (slam) is a vital task for autonomous robots moving in unknown environments. It is a rigorous and complex problem where the signals of different modalities originating from sensors as diverse as laser range finders, IR sensors, ultrasonic transducers, odometers and inertial measurement units are used in its different variants. Expensive sensory arrangements and significant computational resources are desired in all the variants of slam and multi-processing hardware-software arrangements are essentially needed. Consequently, the overall system becomes quite elaborate and expensive. Another common issue is the odometry error;this error accumulates over time and the robot looses track completely or distorts the map. Generally, a notebook computer is used as the computational resource;as it is frequently placed and removed from the robot, the odometry information is significantly affected. Therefore, it is desired to do the sensor calibration with fixed and complete system hardware. In this work, a testbed development for slam is reported. In this design, a low-cost and light-weight FPGA SoC development board is used as the main platform along with a low-cost LIDAR sensor;both fixed on the robot and controlled wirelessly. This fixes the whole hardware configuration and hence the weight of the robot, therefore the calibration results do not alter over time. Moreover, a HW-SW co-design approach is adopted to achieve a highly capable but very compact computational resource. Post-processing of the sensory data with rigorous calibration models has resulted in reasonably precise robot's odometry and environmental perception. The complete system is tested in indoor environment and satisfactory localization and mapping results are achieved.

关键词： slam algorithm Extended Kalman Filter FPGA HW-SW Co-Design

来源：评论

学校读者我要写书评

暂无评论

A Stable and Accurate Marker-less Augmented Reality Registration Method

A Stable and Accurate Marker-less Augmented Reality Registra...

引用

International Conference on Cyberworlds (CW)

作者： Gao, Qing Hong Wan, Tao Ruan Tang, Wen Chen, Long Xian Polytech Univ Coll Elect & Informat Xian Shaanxi Peoples R China Univ Bradford Fac Informat & Engn Bradford W Yorkshire England Bournemouth Univ Fac Sci & Technol Bournemouth Dorset England

ISBN: (纸本)9780769562155

Markerless Augmented Reality (AR) registration using the standard Homography matrix is unstable, and for image-based registration it has very low accuracy. In this paper, we present a new method to improve the stability and the accuracy of marker-less registration in AR. Based on the Visual Simultaneous Localization and Mapping (V-slam) framework, our method adds a three-dimensional dense cloud processing step to the state-of-the-art ORB-slam in order to deal with mainly the point cloud fusion and the object recognition. Our algorithm for the object recognition process acts as a stabilizer to improve the registration accuracy during the model to the scene transformation process. This has been achieved by integrating the Hough voting algorithm with the Iterative Closest Points(ICP) method. Our proposed AR framework also further increases the registration accuracy with the use of integrated camera poses on the registration of virtual objects. Our experiments show that the proposed method not only accelerates the speed of camera tracking with a standard slam system, but also effectively identifies objects and improves the stability of markerless augmented reality applications.

关键词： Augmented Reality slam algorithm Point Clouds Hough voting algorithm

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：