检索结果-内蒙古大学图书馆

LuoJia-Explorer: Unmanned Collaborative Localization and Mapping System 1

3rd International Conference on Autonomous Unmanned Systems, ICAUS 2023

作者： Wu, Weitong Chen, Chi Yang, Bisheng State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing Wuhan University Wuhan430079 China Engineering Research Center of Space-Time Data Capturing and Smart Application The Ministry of Education of P.R.C. Wuhan430079 China

ISBN: (数字)9789819710997

ISBN: (纸本)9789819710980

Over the past two decades, numerous single-robot SLAM systems have been developed and proposed. However, the problem of collaborative localization and mapping with a team of robots remains an open challenge. Collaborative localization and mapping plays a crucial role in enhancing mapping efficiency and has a wide application in search and rescue, infrastructure monitoring, extraterrestrial planetary cave exploration, etc. In this paper, we propose LuoJia-Explorer, an unmanned collaborative localization and mapping system for ground platforms. The proposed system is centralized with a main control center. Each ground platform is equipped with a multi-beam LiDAR and IMU to perform LiDAR-inertial odometry in its own coordinate system. Through mesh networking communication technology, the main control center receives the mapping result and trajectory from each platform. At the control center, global SLAM is performed, which includes loop detection, relative pose estimation and global pose graph optimization, to obtain a globally consistent mapping result. Two ground platforms were built to verify the proposed system. Field experiments have preliminarily verified the effectiveness of this system. © Beijing HIWING Scientific and Technological Information Institute 2024.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

CG-SSD: Corner Guided Single Stage 3D Object Detection from LiDAR Point Cloud

arXiv

引用

arXiv 2022年

作者： Ma, Ruiqi Chen, Chi Yang, Bisheng Li, Deren Wang, Haiping Cong, Yangzi Hu, Zongtian State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing Wuhan University Wuhan430079 China Engineering Research Center of Space-Time Data Capturing and Smart Application The Ministry of Education of P.R.C. China

At present, the anchor-based or anchor-free models that use LiDAR point clouds for 3D object detection use the center assigner strategy to infer the 3D bounding boxes. However, in a real world scene, the LiDAR can only acquire a limited object surface point clouds, but the center point of the object does not exist. Obtaining the object by aggregating the incomplete surface point clouds will bring a loss of accuracy in direction and dimension estimation. To address this problem, we propose a corner-guided anchor-free single-stage 3D object detection model (CG-SSD1).Firstly, 3D sparse convolution backbone network composed of residual layers and sub-manifold sparse convolutional layers are used to construct bird's eye view (BEV) features for further deeper feature mining by a lite U-shaped network;Secondly, a novel corner-guided auxiliary module (CGAM) is proposed to incorporate corner supervision signals into the neural network. CGAM is explicitly designed and trained to detect partially visible and invisible corners to obtains a more accurate object feature representation, especially for small or partial occluded objects;Finally, the deep features from both the backbone networks and CGAM module are concatenated and fed into the head module to predict the classification and 3D bounding boxes of the objects in the scene. The experiments demonstrate CG-SSD achieves the state-of-art performance on the ONCE benchmark for supervised 3D object detection using single frame point cloud data, with 62.77%mAP. Additionally, the experiments on ONCE and Waymo Open dataset show that CGAM can be extended to most anchor-based models which use the BEV feature to detect objects, as a plug-in and bring +1.17%∼+14.27%AP *** Codes 00-02 Copyright © 2022, The Authors. All rights reserved.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

PlaneSAM: Multimodal Plane Instance Segmentation Using the Segment Anything Model

arXiv

引用

arXiv 2024年

作者： Deng, Zhongchen Yang, Zhechen Chen, Chi Zeng, Cheng Meng, Yan Yang, Bisheng School of Computer Science and Information Engineering Hubei University Hubei Wuhan430062 China School of Artificial Intelligence Hubei University Hubei Wuhan430062 China Ministry of Education Hubei Wuhan430062 China State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing Wuhan University Hubei Wuhan430072 China Engineering Research Center of Space-Time Data Capturing and Smart Application The Ministry of Education of P.R.C. Hubei Wuhan430072 China Institute of Geospatial Intelligence Wuhan University Hubei Wuhan430072 China

Plane instance segmentation from RGB-D data is a crucial research topic for many downstream tasks, such as indoor 3D reconstruction. However, most existing deep-learning-based methods utilize only information within the RGB bands, neglecting the important role of the depth band in plane instance segmentation. Based on EfficientSAM, a fast version of the Segment Anything Model (SAM), we propose a plane instance segmentation network called PlaneSAM, which can fully integrate the information of the RGB bands (spectral bands) and the D band (geometric band), thereby improving the effectiveness of plane instance segmentation in a multimodal manner. Specifically, we use a dual-complexity backbone, with primarily the simpler branch learning D-band features and primarily the more complex branch learning RGB-band features. Consequently, the backbone can effectively learn D-band feature representations even when D-band training data is limited in scale, retain the powerful RGB-band feature representations of EfficientSAM, and allow the original backbone branch to be fine-tuned for the current task. To enhance the adaptability of our PlaneSAM to the RGB-D domain, we pretrain our dual-complexity backbone using the segment anything task on large-scale RGB-D data through a self-supervised pretraining strategy based on imperfect pseudo-labels. To support the segmentation of large planes, we optimize the loss function combination ratio of EfficientSAM. In addition, Faster R-CNN is used as a plane detector, and its predicted bounding boxes are fed into our dual-complexity network as prompts, thereby enabling fully automatic plane instance segmentation. Experimental results show that the proposed PlaneSAM sets a new state-of-the-art (SOTA) performance on the ScanNet dataset, and outperforms previous SOTA approaches in zero-shot transfer on the 2D-3D-S, Matterport3D, and ICL-NUIM RGB-D datasets, while only incurring a 10% increase in computational overhead compared to EfficientSAM. Ou

关键词： 3D reconstruction

来源：评论

学校读者我要写书评

暂无评论

Survey of reduced workload and crewing strategies for ocean patrol vessels

引用

NAVAL ENGINEERS JOURNAL 2001年第2期113卷 29-44页

作者： Baker, C Krull, R Snyder, G Lincoln, W Malone, TB Clifford C. Baker CIE CHFEP is a senior staff scientist at Carlow International Incorporated. He has applied most of his 24 years of experience in the application of human engineering technology to maritime systems. Mr. Baker has directed much of Carlow's efforts to reduce ship workload and to improve human performance and maritime safety through application of human factors methods and data. He is a Certified Industrial Ergonomist (CIE) as well as a Certified Human Factors Engineering Professional (CHFEP). Both certifications were granted by Oxford Research where Mr. Baker also serves as an Advisory Board member. Russell D. Krull P.E. is a senior engineer with A&T/Proteus Engineering with more than 18 years of experi-ence in marine engineering naval architecture and program management including 16 years of active duty in the U.S. Coast Guard. Recent experience includes advanced ship design studies engineering software development technical support for the USMC Advanced Amphibious Assault Vehicle propulsion systems analyses ship structural engineering and cargo handling systems engineering. Mr. Krull has an M.S.E. in naval architecture and marine engineering and an M.S.E. in industrial and operations engineering from University of Michigan and a B.S. in ocean engineering from the U.S. Coast Guard Academy. Capt. Glenn L. Snyder USCG. Regrettably since this paper was originally written Capt. Snyder has passed away. At the time of his death he was an operations specialist assigned to the Coast Guard's Deepwater Capabilities Replacement Project as Chief of Human Systems Integration. He served as commanding officer of the patrol boat Cape George (WPB-95306) the icebreaking tug Biscayne Bay (WTGB-104) and the cutter Legare (WMEC-911). A 1975 graduate of the U.S. Coast Guard Academy Capt. Snyder held an M.A. in national security and strategic studies from the U.S. Naval War College and an M.A. in international relations from Salve Regina College. In addition he was a 1998 fellow of the Foreign Service

The U.S. Coast Guard is in the concept exploration phase of its Integrated Deepwater System (IDS) acquisition project. This project will define the next generation of surface, air and command, control, communications, computers, intelligence, surveillance, and reconnaissance (C4ISR) assets used to perform the Coast Guard's missions in the IDS environment (>50 NM off the U. S. coastline). As part of early technology investigations, the needs exist to: (1) analyze the workload requirements of the IDS, (2) identify alternative means to perform ship's work, and (3) optimize ship manning consistent with ship workload, performance criteria, and the available tools and equipment aboard. To reduce shipboard work requires an understanding of the mission and support requirements placed on the vessel and crew, how these requirements are currently met, and how requirements might otherwise be met to reduce workload and crew size. This study examined currently implemented workload and manpower reducing approaches of commercial maritime fleets, U.S. and foreign navies, and foreign coastguards. These approaches were analyzed according to evaluation criteria approved by the IDS acquisition project team. From this, strategies for shipboard work reduction that may be considered for adoption by the IDS were identified and analyzed according to performance and costs factors. Ten workload-reducing strategies were identified: damage control, bridge, multiple crewing, engineering, risk acceptance, modularity, deck, enabling technologies, ship/personnel readiness, and operability and maintainability.

关键词： Naval vessels

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：