[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84158-en":3,"doc-seo-84158-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},84158,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Dynamic Object Detection and Tracking in Construction LiDAR and Fisheye Sensor Fusion Model","Robust dynamic object detection and tracking enables quadruped robots to operate safely with humans in construction environments. Building on LiDAR SLAM and occupancy grid motion cues, the framework detects moving objects in a registered point cloud and projects their 3D coordinates onto a 2D cylindrical panorama from an upward-facing fisheye camera for semantic labeling. Semantic updates guide observation integration into a modified Kalman filter for improved state estimation, maintaining high precision and robustness when objects switch between dynamic and temporarily static conditions.","Dynamic Object Detection and Tracking in Construction: A Fisheye Camera and LiDAR Sensor Fusion Model  \nYilong Chen, Huili Huang, and Yong K. Cho  \nAbstract--Robust dynamic object detection and tracking are essential for enabling robots to operate safely and effectively alongside humans in complex environments such as construction sites. While LiDAR-based SLAM and occupancy grid methods offer viable solutions for detecting and tracking motion, many state-of-the-art 3D vision approaches rely heavily on pre-trained neural networks and require additional post-processing to identify moving objects. Sensor fusion techniques, combining the precision of LiDAR with the semantic richness of RGB imagery, offer a promising alternative. In this work, we present a novel framework that enhances a quadruped robot equipped with a LiDAR sensor and an upward-facing fisheye camera for realtime dynamic object detection and tracking. After identifying moving objects within a registered point cloud, our method assigns semantic labels by projecting 3D coordinates onto a 2D cylindrical panorama, aligning with real-time image-based detections for observation update of the Kalman filter. The proposed system demonstrates high precision, simplicity, and robustness, particularly in handling objects transitioning between dynamic and static states, thus it is well-suited for deployment in real-world construction environments.  \nI. INTRODUCTION  \nSimultaneous Localization and Mapping (SLAM) has been widely adopted for legged robots equipped with onboard sensors to reconstruct both indoor [1] and outdoor [2] environments, particularly in response to the growing integration of robotic systems in construction tasks. However, dynamic objects are often treated as noise in the SLAM problem, leading most existing approaches to separate dynamic object tracking from map reconstruction [3] . Despite this, discrepancies between consecutive point cloud maps can offer critical insights into object motion relative to the sensor, presenting an opportunity to integrate perception and mapping more effectively. In prior work [4], we proposed an online dynamic object detection and tracking method based on LiDAR SLAM and occupancy grids. By assigning discrete states to grid cells and updating occupancy probabilities through discounted returns across state transitions, dynamic objects can be clustered. This information is then fed into a Kalman filter for tracking and motion state estimation, without requiring any additional sensing beyond the LiDAR.  \nNevertheless, two key challenges persist: (1) the inability to semantically identify detected moving objects, and (2) degraded tracking performance when objects transition from dynamic to temporarily static states. While incorporating a camera can address these issues by providing semantic context, existing deep learning-based 3D object detection approaches often demand complex models, substantial computational resources, extensive labeled datasets, and heavy  \nYC, HH, YC are with Georgia Institute of Technology, Atlanta, GA 30332, USA, {ychen3339, [hhuang413}@gatech.edu](hhuang413}@gatech.edu), [yong.cho@ce.gatech.edu](yong.cho@ce.gatech.edu)  \nFigure 1. Frame System (top) and Point Projection and Mapping (bottom).  \ndependence on accurate depth estimation [5], which can be error-prone. Conventional RGB cameras also suffer from limited fields of view, and using multiple cameras introduces further challenges in extrinsic and temporal calibration [6] . Additionally, distortion correction is required when using pinhole lens models. To address these limitations, we integrate a single upward-facing fisheye camera into our existing LiDAR-based system, enabling a unified and semantically rich perception framework.  \nThis research proposes a novel method for dynamic object detection and tracking by combining registered LiDAR point cloud maps from SLAM with cylindrical panoramic images derived from the fisheye camera (Fig. 1) . The framew","cbCaitNeBYzrd5Ye","https://ap.wps.com/l/cbCaitNeBYzrd5Ye","pdf",378587,1,4,"English","en",105,"# Introduction\n## Motivation and limitations of existing SLAM approaches\n## Proposed LiDAR–fisheye fusion framework and contributions\n# Related Work\n## LiDAR-camera fusion and 3D detection with semantic/transformer methods","[{\"question\":\"Why is dynamic object detection and tracking important for construction-site robots?\",\"answer\":\"Robots must perceive moving objects to operate safely and effectively around humans in complex construction environments. Detecting motion and tracking object states directly supports reliable navigation in changing scenes.\"},{\"question\":\"How does the model fuse LiDAR and the fisheye camera for semantic tracking?\",\"answer\":\"The approach detects moving objects in a registered LiDAR point cloud, then assigns semantic labels by projecting 3D coordinates onto a 2D cylindrical panorama derived from the upward-facing fisheye camera. These semantic observations are integrated into the tracking update of a modified Kalman filter.\"},{\"question\":\"What tracking challenge does the proposed system address during dynamic-to-static transitions?\",\"answer\":\"Tracking performance degrades when objects move and then become temporarily static. The method improves robustness by combining occupancy-grid-based motion cues with semantic augmentation from the fused camera-LiDAR pipeline.\"}]",1784193517,10,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"dynamic-object-detection-and-tracking-in-construction-lidar-and-fisheye-sensor-fusion-model","",{"@graph":35,"@context":84},[36,52,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":21},"https://docshare.wps.com/document/dynamic-object-detection-and-tracking-in-construction-lidar-and-fisheye-sensor-fusion-model/84158/",{"url":51,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":23,"description":14,"dateModified":60,"datePublished":61,"encodingFormat":59,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":40,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":20},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"Why is dynamic object detection and tracking important for construction-site robots?","Question",{"text":74,"@type":75},"Robots must perceive moving objects to operate safely and effectively around humans in complex construction environments. Detecting motion and tracking object states directly supports reliable navigation in changing scenes.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does the model fuse LiDAR and the fisheye camera for semantic tracking?",{"text":79,"@type":75},"The approach detects moving objects in a registered LiDAR point cloud, then assigns semantic labels by projecting 3D coordinates onto a 2D cylindrical panorama derived from the upward-facing fisheye camera. These semantic observations are integrated into the tracking update of a modified Kalman filter.",{"name":81,"@type":72,"acceptedAnswer":82},"What tracking challenge does the proposed system address during dynamic-to-static transitions?",{"text":83,"@type":75},"Tracking performance degrades when objects move and then become temporarily static. The method improves robustness by combining occupancy-grid-based motion cues with semantic augmentation from the fused camera-LiDAR pipeline.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,127,130,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":28,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":28,"slug":132},"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]