{"title":"基于车辆参考的低信道路边激光雷达点云注册","authors":"Ciyun Lin;Shuangjie Deng;Bowen Gong;Hongchao Liu","doi":"10.1109/JSEN.2024.3496782","DOIUrl":null,"url":null,"abstract":"The light detection and ranging (LiDAR) sensor has demonstrated its preponderance in high-resolution, lane-level traffic object identification, and trajectory tracking. It empowers road users, such as vehicles, pedestrians, and cyclists, by notifying them of dangerous behaviors and potential collisions. However, occlusion and consecutive tracking remain significant challenges in single roadside LiDAR detection. One promising solution to address this problem involves deploying roadside LiDAR sensors at different locations along the road. In such scenarios, point cloud registration becomes a crucial task in processing point cloud data. This article presents a novel point cloud registration method for low-channel roadside LiDAR, utilizing the moving vehicle as the reference points. First, ground normal vectors were introduced to rotate point clouds and achieve z-axis height alignment. Second, the motion vehicle was extracted and employed as the reference point. A 2-D reference point correction algorithm was developed to enhance the identification accuracy of the reference point. Third, a two-step registration framework was proposed, using the 2-D reference points for partial registration. The iterative closest point (ICP) algorithm was employed for accurate global registration. Experimental evaluations are conducted using point cloud data collected from diverse traffic scenarios to assess the proposed method. The results demonstrated significant improvements in data registration accuracy compared to previous point cloud registration algorithms. The relative rotation error (RRE) and relative translation error (RTE) were less than 1° and 0.5 m, respectively.","PeriodicalId":447,"journal":{"name":"IEEE Sensors Journal","volume":"25 1","pages":"1198-1208"},"PeriodicalIF":4.3000,"publicationDate":"2024-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Vehicle Reference-Based Point Cloud Registration for Low-Channel Roadside LiDARs\",\"authors\":\"Ciyun Lin;Shuangjie Deng;Bowen Gong;Hongchao Liu\",\"doi\":\"10.1109/JSEN.2024.3496782\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The light detection and ranging (LiDAR) sensor has demonstrated its preponderance in high-resolution, lane-level traffic object identification, and trajectory tracking. It empowers road users, such as vehicles, pedestrians, and cyclists, by notifying them of dangerous behaviors and potential collisions. However, occlusion and consecutive tracking remain significant challenges in single roadside LiDAR detection. One promising solution to address this problem involves deploying roadside LiDAR sensors at different locations along the road. In such scenarios, point cloud registration becomes a crucial task in processing point cloud data. This article presents a novel point cloud registration method for low-channel roadside LiDAR, utilizing the moving vehicle as the reference points. First, ground normal vectors were introduced to rotate point clouds and achieve z-axis height alignment. Second, the motion vehicle was extracted and employed as the reference point. A 2-D reference point correction algorithm was developed to enhance the identification accuracy of the reference point. Third, a two-step registration framework was proposed, using the 2-D reference points for partial registration. The iterative closest point (ICP) algorithm was employed for accurate global registration. Experimental evaluations are conducted using point cloud data collected from diverse traffic scenarios to assess the proposed method. The results demonstrated significant improvements in data registration accuracy compared to previous point cloud registration algorithms. The relative rotation error (RRE) and relative translation error (RTE) were less than 1° and 0.5 m, respectively.\",\"PeriodicalId\":447,\"journal\":{\"name\":\"IEEE Sensors Journal\",\"volume\":\"25 1\",\"pages\":\"1198-1208\"},\"PeriodicalIF\":4.3000,\"publicationDate\":\"2024-11-18\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"IEEE Sensors Journal\",\"FirstCategoryId\":\"103\",\"ListUrlMain\":\"https://ieeexplore.ieee.org/document/10756581/\",\"RegionNum\":2,\"RegionCategory\":\"综合性期刊\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENGINEERING, ELECTRICAL & ELECTRONIC\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Sensors Journal","FirstCategoryId":"103","ListUrlMain":"https://ieeexplore.ieee.org/document/10756581/","RegionNum":2,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}
Vehicle Reference-Based Point Cloud Registration for Low-Channel Roadside LiDARs
The light detection and ranging (LiDAR) sensor has demonstrated its preponderance in high-resolution, lane-level traffic object identification, and trajectory tracking. It empowers road users, such as vehicles, pedestrians, and cyclists, by notifying them of dangerous behaviors and potential collisions. However, occlusion and consecutive tracking remain significant challenges in single roadside LiDAR detection. One promising solution to address this problem involves deploying roadside LiDAR sensors at different locations along the road. In such scenarios, point cloud registration becomes a crucial task in processing point cloud data. This article presents a novel point cloud registration method for low-channel roadside LiDAR, utilizing the moving vehicle as the reference points. First, ground normal vectors were introduced to rotate point clouds and achieve z-axis height alignment. Second, the motion vehicle was extracted and employed as the reference point. A 2-D reference point correction algorithm was developed to enhance the identification accuracy of the reference point. Third, a two-step registration framework was proposed, using the 2-D reference points for partial registration. The iterative closest point (ICP) algorithm was employed for accurate global registration. Experimental evaluations are conducted using point cloud data collected from diverse traffic scenarios to assess the proposed method. The results demonstrated significant improvements in data registration accuracy compared to previous point cloud registration algorithms. The relative rotation error (RRE) and relative translation error (RTE) were less than 1° and 0.5 m, respectively.
期刊介绍:
The fields of interest of the IEEE Sensors Journal are the theory, design , fabrication, manufacturing and applications of devices for sensing and transducing physical, chemical and biological phenomena, with emphasis on the electronics and physics aspect of sensors and integrated sensors-actuators. IEEE Sensors Journal deals with the following:
-Sensor Phenomenology, Modelling, and Evaluation
-Sensor Materials, Processing, and Fabrication
-Chemical and Gas Sensors
-Microfluidics and Biosensors
-Optical Sensors
-Physical Sensors: Temperature, Mechanical, Magnetic, and others
-Acoustic and Ultrasonic Sensors
-Sensor Packaging
-Sensor Networks
-Sensor Applications
-Sensor Systems: Signals, Processing, and Interfaces
-Actuators and Sensor Power Systems
-Sensor Signal Processing for high precision and stability (amplification, filtering, linearization, modulation/demodulation) and under harsh conditions (EMC, radiation, humidity, temperature); energy consumption/harvesting
-Sensor Data Processing (soft computing with sensor data, e.g., pattern recognition, machine learning, evolutionary computation; sensor data fusion, processing of wave e.g., electromagnetic and acoustic; and non-wave, e.g., chemical, gravity, particle, thermal, radiative and non-radiative sensor data, detection, estimation and classification based on sensor data)
-Sensors in Industrial Practice