Roadside Light Detection And Ranging Research Articles

Near-crash risk identification and evaluation for takeout delivery motorcycles using roadside LiDAR

The proliferation of motorcycles in urban areas has raised concerns regarding traffic safety. However, traditional sensors struggle to obtain precise high-resolution trajectory data, which hinder the accurate identification and quantification of near-crash risks for takeout delivery motorcycles. To fill this gap, this study presents a novel approach utilizing roadside light detection and ranging (LiDAR) to identify and evaluate the risk of near crashes of takeout delivery motorcycles. First, a trajectory amendment method incorporating speed and steering angle was introduced to enhance the accuracy and continuity of the trajectory prediction. Second, a trajectory prediction method combining the steering intention and a repulsive force model was proposed for near-crash risk prediction. Subsequently, a near-crash identification method was developed that relied on the closest distance and risk radius. Finally, near-crash risk fields were constructed to quantify risk levels by leveraging velocity, position, and weight. The experimental results demonstrated 92.10 % accuracy in intention prediction, with mean absolute error (MAE) and root mean square error (RMSE) values of 0.53 m and 0.45 m, respectively. In addition to its higher accuracy, the proposed method makes it easier to quantify near-crash risk and supports a proactive approach for visualizing and analyzing traffic safety.

Vehicle Detection and Tracking with Roadside LiDAR Using Improved ResNet18 and the Hungarian Algorithm.

By the end of the 2020s, full autonomy in autonomous driving may become commercially viable in certain regions. However, achieving Level 5 autonomy requires crucial collaborations between vehicles and infrastructure, necessitating high-speed data processing and low-latency capabilities. This paper introduces a vehicle tracking algorithm based on roadside LiDAR (light detection and ranging) infrastructure to reduce the latency to 100 ms without compromising the detection accuracy. We first develop a vehicle detection architecture based on ResNet18 that can more effectively detect vehicles at a full frame rate by improving the BEV mapping and the loss function of the optimizer. Then, we propose a new three-stage vehicle tracking algorithm. This algorithm enhances the Hungarian algorithm to better match objects detected in consecutive frames, while time-space logicality and trajectory similarity are proposed to address the short-term occlusion problem. Finally, the system is tested on static scenes in the KITTI dataset and the MATLAB/Simulink simulation dataset. The results show that the proposed framework outperforms other methods, with F1-scores of 96.97% and 98.58% for vehicle detection for the KITTI and MATLAB/Simulink datasets, respectively. For vehicle tracking, the MOTA are 88.12% and 90.56%, and the ID-F1 are 95.16% and 96.43%, which are better optimized than the traditional Hungarian algorithm. In particular, it has a significant improvement in calculation speed, which is important for real-time transportation applications.

High-accurate vehicle trajectory extraction and denoising from roadside LIDAR sensors

In recent years, Light Detection and Ranging (LIDAR) has been widely used in the Intelligent Traffic System due to its high-precision data acquisition capabilities. It has great potential for accurately detecting and tracking vehicle trajectories on the roadside. This study aims to develop a novel methodological framework for accurate vehicle trajectory extraction from roadside LIDAR. The proposed method starts by designing a detection area, followed by the application of statistical filtering, density clustering, and a random sample consensus (RANSAC) segmentation algorithm to eliminate the background point cloud. Subsequently, a modified density-based spatial clustering of applications with a noise algorithm (BSO-DBSCAN) is employed to detect the vehicle based on a beetle swarm optimization algorithm. An oriented bounding box (OBB) is used to obtain vehicle position, length, and width. A modified trajectory association method based on the Kalman filtering algorithm is proposed to address vehicle occlusion. LIDAR is deployed to obtain vehicle operation data in the expressway work zone, and the performance of the proposed method is tested using an unmanned aerial vehicle (UAV). The experimental results demonstrate that the proposed method successfully extracts high-precision vehicle trajectories. Specifically, the vehicle recognition accuracy increased by 12.7% in MAPE and 14.9% in RMSE compared to DBSCAN. The trajectory tracking accuracy increased by 13.5%, and the number of ID switching (ID-SW) was reduced by 271 times compared to SORT. The vehicle trajectory data extracted in this study provides a foundation for traffic characteristic analysis and traffic modeling. The extracted data can be downloaded from the following GitHub: https://github.com/gao0628/Dataset.

Research on an Adaptive Method for the Angle Calibration of Roadside LiDAR Point Clouds.

Light Detection and Ranging (LiDAR), a laser-based technology for environmental perception, finds extensive applications in intelligent transportation. Deployed on roadsides, it provides real-time global traffic data, supporting road safety and research. To overcome accuracy issues arising from sensor misalignment and to facilitate multi-sensor fusion, this paper proposes an adaptive calibration method. The method defines an ideal coordinate system with the road's forward direction as the X-axis and the intersection line between the vertical plane of the X-axis and the road surface plane as the Y-axis. This method utilizes the Kalman filter (KF) for trajectory smoothing and employs the random sample consensus (RANSAC) algorithm for ground fitting, obtaining the projection of the ideal coordinate system within the LiDAR system coordinate system. By comparing the two coordinate systems and calculating Euler angles, the point cloud is angle-calibrated using rotation matrices. Based on measured data from roadside LiDAR, this paper validates the calibration method. The experimental results demonstrate that the proposed method achieves high precision, with calculated Euler angle errors consistently below 1.7%.

Center-Aware 3D Object Detection with Attention Mechanism Based on Roadside LiDAR

Infrastructure 3D Object Detection is a pivotal component of Vehicle-Infrastructure Cooperated Autonomous Driving (VICAD). As turning objects account for a high proportion of traffic at intersections, anchor-free representation in the bird’s-eye view (BEV) is more suitable for roadside 3D detection. In this work, we propose CetrRoad, a simple yet effective center-aware detector with transformer-based detection head for roadside 3D object detection with single LiDAR (Light Detection and Ranging). CetrRoad firstly utilizes a voxel-based roadside LiDAR feature encoder module that voxelizes and projects the raw point cloud into BEV with dense feature representation, following a one-stage center proposal module that initializes center candidates of objects based on the top N points in the BEV target heatmap with unnormalized 2D Gaussian. Then, taking attending center proposals as query embedding, a detection head with multi-head self-attention and multi-scale multi-head deformable cross attention can refine and predict 3D bounding boxes for different classes moving/parked at the intersection. Extensive experiments and analyses demonstrate that our method achieves state-of-the-art performance on the DAIR-V2X-I benchmark with an acceptable training time cost, especially for Car and Cyclist. CetrRoad also reaches comparable results with the multi-modal fusion method for Pedestrian. An ablation study demonstrates that center-aware query as input can provide denser supervision than a purified feature map in the attention-based detection head. Moreover, we were able to intuitively observe that in complex traffic environment, our proposed model could produce more accurate 3D detection results than other compared methods with fewer false positives, which is helpful for other downstream VICAD tasks.

3D Point Cloud Stitching for Object Detection with Wide FoV Using Roadside LiDAR

Light Detection and Ranging (LiDAR) is widely used in the perception of physical environment to complete object detection and tracking tasks. The current methods and datasets are mainly developed for autonomous vehicles, which could not be directly used for roadside perception. This paper presents a 3D point cloud stitching method for object detection with wide horizontal field of view (FoV) using roadside LiDAR. Firstly, the base detection model is trained by KITTI dataset and has achieved detection accuracy of 88.94. Then, a new detection range of 180° can be inferred to break the limitation of camera’s FoV. Finally, multiple sets of detection results from a single LiDAR are stitched to build a 360° detection range and solve the problem of overlapping objects. The effectiveness of the proposed approach has been evaluated using KITTI dataset and collected point clouds. The experimental results show that the point cloud stitching method offers a cost-effective solution to achieve a larger FoV, and the number of output objects has increased by 77.15% more than the base model, which improves the detection performance of roadside LiDAR.

Time Synchronization and Space Registration of Roadside LiDAR and Camera

The sensing system consisting of Light Detection and Ranging (LiDAR) and a camera provides complementary information about the surrounding environment. To take full advantage of multi-source data provided by different sensors, an accurate fusion of multi-source sensor information is needed. Time synchronization and space registration are the key technologies that affect the fusion accuracy of multi-source sensors. Due to the difference in data acquisition frequency and deviation in startup time between LiDAR and the camera, asynchronous data acquisition between LiDAR and camera is easy to occur, which has a significant influence on subsequent data fusion. Therefore, a time synchronization method of multi-source sensors based on frequency self-matching is developed in this paper. Without changing the sensor frequency, the sensor data are processed to obtain the same number of data frames and set the same ID number, so that the LiDAR and camera data correspond one by one. Finally, data frames are merged into new data packets to realize time synchronization between LiDAR and camera. Based on time synchronization, to achieve spatial synchronization, a nonlinear optimization algorithm of joint calibration parameters is used, which can effectively reduce the reprojection error in the process of sensor spatial registration. The accuracy of the proposed time synchronization method is 99.86% and the space registration accuracy is 99.79%, which is better than the calibration method of the Matlab calibration toolbox.

Objects Detection With 3-D Roadside LIDAR Under Snowy Weather

Rain and snowfall will increase noise, change the resolution of objects in the point cloud and present great challenges to the accurate recognition of traffic objects. Accordingly, this article proposes an efficient real-time method for roadside 3-D light detection and ranging (LIDAR) background point cloud extraction and object segmentation under snowy weather. We first use a historical point cloud sequence to quickly construct a background model, extract the background point cloud from the current frame by using a background difference method and update the background model in real-time. Then, the noise caused by snowfall in the non-background point cloud is filtered based on the beam density difference of the object point clouds. Finally, the remaining object point cloud is accurately segmented based on the proposed hierarchical object clustering method. We use an intelligent roadside system equipped with 3-D roadside LIDAR to collect point cloud data in a snowfall environment and evaluate the proposed method qualitatively and quantitatively. Experimental results show that the proposed method effectively avoids the problem of under-segmentation and over-segmentation of object point clouds under snowy conditions, and the precision and recall rate of traffic object segmentation reached 96.41% and 95.02%, respectively, thereby indicating a significant improvement in the accuracy and reliability of traffic object detection using roadside LIDAR under snowy conditions.

Object Detection Based on Roadside LiDAR for Cooperative Driving Automation: A Review

Light Detection and Ranging (LiDAR) technology has the advantages of high detection accuracy, a wide range of perception, and not being affected by light. The 3D LiDAR is placed at the commanding height of the traffic scene, the overall situation can be grasped from the perspective of top view, and the trajectory of each object in the traffic scene can be accurately perceived in real time, and then the object information can be distributed to the surrounding vehicles or other roadside LiDAR through advanced wireless communication equipment, which can significantly improve the local perception ability of an autonomous vehicle. This paper first describes the characteristics of roadside LiDAR and the challenges of object detection and then reviews in detail the current methods of object detection based on a single roadside LiDAR and multi-LiDAR cooperatives. Then, some studies for roadside LiDAR perception in adverse weather and datasets released in recent years are introduced. Finally, some current open challenges and future works for roadside LiDAR perception are discussed. To the best of our knowledge, this is the first work to systematically study roadside LiDAR perception methods and datasets. It has an important guiding role in further promoting the research of roadside LiDAR perception for practical applications.

A Novel Information Theory-Based Metric for Evaluating Roadside LiDAR Placement

Intelligent vehicle infrastructure cooperative systems can essentially address the bottleneck of the limited perception range of autonomous vehicles. This is achieved by the real-time perception of road traffic using roadside mounted sensors, which is receiving increasing attention. Light detection and ranging (LiDAR) has progressively been deployed on the roadside due to its advantage of remaining unaffected by lights and the capability to obtain 3-D information about traffic participants. However, a practical evaluation criterion is currently lacking for assessing the merits of roadside LiDAR placement. In this study, we propose an information-theoretic LiDAR mounting placement evaluation method, which simplifies the complex problem of LiDAR mounting placement to a simple parameter calculation problem. We characterize the amount of information detected by LiDAR in the region of interest (ROI) by introducing the per voxel point clouds’ metric. The resulting metric, named the maximum density gain, quantifies the effect of LiDAR mounting placements on perception performance. A bidirectional eight-lane traffic scenario is constructed in Unity3D, and datasets of simulated LiDARs are collected under two traffic conditions: with and without obstruction. The Pointpillars is used to test the collected dataset, which is a point cloud-based 3-D object detection algorithm. The results of Pointpillars and our proposed method are compared to test the validity of the proposed method. The experimental results show that our proposed method is compatible with the outcomes of Pointpillars while being simpler and computationally efficient. The proposed LiDAR mounting placement evaluation method can offer a reliable calculation reference for the mounting placement of roadside LiDARs.

Leveraging Deep Convolutional Neural Networks Pre-Trained on Autonomous Driving Data for Vehicle Detection From Roadside LiDAR Data

Recent technological advancements in computer vision algorithms and data acquisition devices have greatly facilitated the research and applications of deep learning-based traffic object recognition from Light Detection and Ranging (LiDAR) data. The majority of existing methodologies applied deep learning (DL)-based techniques, especially Convolutional Neural Networks (CNNs), for vehicle detection and tracking on autonomous driving datasets. Nevertheless, fewer studies were focused on DL-based vehicle detection using roadside LiDAR data, partially due to the lack of publicly available roadside LiDAR datasets for network training and testing. This paper develops a novel framework based on CNNs and LiDAR data for automated vehicle detection. It leverages the domain knowledge of CNNs trained on large-scale autonomous driving datasets for vehicle detection from roadside LiDAR data. In the experimental study, roadside LiDAR data were collected at a road intersection in Reno, Nevada, U.S. Meanwhile, a CNN architecture was proposed to detect vehicles from LiDAR data through 3D bounding boxes. The proposed CNN was modified from the established PointPillars network by adding dense connections to the convolutional layers to achieve more comprehensive feature extraction. Three CNNs, including the proposed CNN, PointPillars, and YOLOv4, were trained and tested on PandaSet, a publicly available large-scale autonomous driving LiDAR dataset. Subsequently, the trained CNNs were reused for vehicle detection from the captured roadside LiDAR data. The experimental results demonstrated that the proposed CNN outperformed the others in the testing metrics. All three networks showed good performance on vehicle detection from the captured roadside LiDAR data.

Traffic Sign Based Point Cloud Data Registration with Roadside LiDARs in Complex Traffic Environments

The intelligent road is an important component of the intelligent vehicle infrastructure cooperative system, the latest development of intelligent transportation systems. As an advanced sensor, Light Detection and Ranging (LiDAR) has gradually been used to collect high-resolution micro-traffic data on the roadside of intelligent roads. Furthermore, a fusion of multiple LiDARs has become a current hot spot to extend the data collection range and improve detection accuracy. This paper focuses on point cloud registration in a complex traffic environment and proposes a three-dimensional (3D) registration method based on traffic signs and prior knowledge of traffic scenes. Traffic signs with their reflective films are used as reference targets to register 3D point cloud data from roadside LiDARs. The proposed method consists of a vertical registration and a horizontal registration. For the vertical registration, we propose a panel rotation algorithm to rotate the initial point cloud to register it vertically, converting the 3D point cloud registration into a two-dimensional (2D) rigid body transformation. For the vertical registration, our system registers traffic signs from different LiDARs. Our method has been verified in some actual scenarios. Compared with previous methods, the proposed method is automatic and does not need to search reference targets manually. Furthermore, it is suitable for actual engineering use and can be applied to sparse point cloud data from LiDAR with few beams, realizing point cloud registration of large disparity.

Fast-Spherical-Projection-Based Point Cloud Clustering Algorithm

Roadside LiDAR (light detection and ranging) is a solution to fill in the gaps for connected vehicles (CV) by detecting the status of global road users at transportation facilities. It relies greatly on the clustering algorithm for accurate and rapid data processing so as to ensure effectiveness and reliability. To contribute to better roadside LiDAR-based transportation facilities, this paper presents a fast-spherical-projection-based clustering algorithm (FSPC) for real-time LiDAR data processing with higher clustering accuracy and noise handling. The FSPC is designed to work on a spherical map which could be directly derived from the instant returns of a LiDAR sensor. A 2D-window searching strategy is specifically designed to accelerate the computation and alleviate the density variation impact in the LiDAR point cloud. The test results show the proposed algorithm can achieve a high processing efficiency with 24.4 ms per frame, satisfying the real-time requirement for most common LiDAR applications (100 ms per frame), and it also ensures a high accuracy in object clustering, with 96%. Additionally, it is observed that the proposed FSPC allows a wider detection range and is more stable, tackling the surge in foreground points that frequently occurs in roadside LiDAR applications. Finally, the generality of the proposed FSPC indicates the proposed algorithm could also be implemented in other areas such as autonomous driving and remote sensing.

A dense background representation method for traffic surveillance based on roadside LiDAR

The performance of traditional video-based traffic surveillance systems is susceptible to illumination variation and perspective distortion. This has been a significant motivation in recent years for research into Light Detection and Ranging (LiDAR)-based traffic surveillance systems, as LiDAR is insensitive to both factors. The first step in LiDAR data processing involves effective extraction of moving foreground objects from a referenced background. However, existing methods only detect a static background based on LiDAR point density or relative distance. In this research, we develop a novel dense background representation model (DBRM) for stationary roadside LiDAR sensors to detect both static and dynamic backgrounds, for freeway traffic surveillance purposes. Background objects tend to be stationary in space and time. DBRM utilizes this property to detect two types of background: both static and dynamic. While the static background is represented by fixed structures, the dynamic background – which may be characterized by quasi-static objects such as tree foliage – is modeled by mixtures of Gaussian probability distributions. Experiments were carried out in two different scenarios to compare the proposed model with two other state-of-the-art models. The results demonstrate the effectiveness, robustness, and detail-preserving advantages of the proposed model.

Automatic Identification of Vehicle Partial Occlusion in Data Collected by Roadside LiDAR Sensors

Light detection and ranging (LiDAR) sensors are receiving an increasing amount of attention in traffic detection because of their powerful capacity for providing accurate trajectory data of vehicles and non-motorized road users. When installed at the roadside, LiDAR faces the same occlusion problem as other over-roadway sensors (such as video cameras)—the integrity and reliability of object detection can be reduced when occlusion occurs. Existing occlusion reasoning methods are either developed for video sensors or require sensor fusion. Fast and accurate occlusion determination is an indispensable step in the process of vehicle detection, classification, and tracking. This paper presents a novel algorithm to automatically identify the occurrence of vehicle partial occlusion and the corresponding occluding/occluded relationship from roadside LiDAR data. According to the inherent characteristics of LiDAR sensors, two specifically designed heatmaps (ClusterID heatmap and Distance heatmap) were generated and used as the basis for fast occlusion search. Using the field data collected from two testbeds, it was shown that the proposed method could achieve about 95.60% accuracy for partial occlusion identification and a processing speed of 2 ms per frame (32-laser LiDAR sensor with 0.1 s per frame).

Grid-based lane identification with roadside LiDAR data

Lane identification is important for many different applications, especially for connected-vehicle technologies. This paper presents a new method for lane identification with the roadside light detection and ranging (LiDAR) serving connected-vehicles. The proposed lane identification method is a revised grid-based clustering method (RGBC). The whole procedure includes background filtering, object clustering, object classification, and RGBC. A location matrix (LM) can be generated to store the location of each lane. The performance of the proposed method was evaluated with the data collected from the real world. The testing results showed that the RGBC can locate 96.73% of vehicles to the correct lane. The RGBC was also compared to the state of the art, showing that the computational load for RGBC is lowest compared to other algorithms, with a cost of slightly reduced accuracy. The time delay for real-time data processing is less than 0.1 ms, which can provide the high-resolution micro traffic data (HRMTD) for connected-vehicles.

Augmented Multiple Vehicles’ Trajectories Extraction Under Occlusions With Roadside LiDAR Data

Object occlusion is a common issue in Light Detection and Ranging (LiDAR)-based vehicle tracking technology. The occlusions can cause variance in vehicle location and speed calculation. How to link the vehicle trajectories caused by occlusion issues is a challenge for traffic engineers and researchers. This paper developed an augmented vehicle tracking method under occlusions with the roadside LiDAR data. The proposed method can be divided into two parts. The first part based on the corner point is used to choose a representative vehicle tracking point. And the second part based on the GNN algorithm is employed to link the vehicles’ trajectories under two occlusion situations. The performance of the proposed method has been evaluated using roadside lidar data collected from four different scenarios. The test results showed that more than 89% of disconnected trajectories can be fixed with the proposed method, which is superior compared to the state-of-the-art method. The proposed method can benefit a lot of transportation areas, such as traffic volume count, vehicle speed tracking, and traffic safety analysis.

Object Classification with Roadside LiDAR Data Using a Probabilistic Neural Network

Object classification is important information for different transportation areas. This research developed a probabilistic neural network (PNN) classifier for object classification using roadside Light Detection and Ranging (LiDAR). The objective was to classify the road user on the urban road into one of four classes: Pedestrian, bicycle, passenger car, and truck. Five features calculated from the point cloud generated from the roadside LiDAR were selected to represent the difference between different classes. A total of 2736 records (2062 records for training, and 674 records for testing) were manually marked for training and testing the PNN algorithm. The data were collected at three different sites representing different scenarios. The performance of the classification was evaluated by comparing the result of the PNN with those of the support vector machine (SVM) and the random forest (RF). The comparison results showed that the PNN can provide the results of classification with the highest accuracy among the three investigated methods. The overall accuracy of the PNN for object classification was 97.6% using the testing database. The errors in the classification results were also diagnosed. Discussions about the direction of future studies were also provided at the end of this paper.

An Automatic Lane Marking Detection Method With Low-Density Roadside LiDAR Data

Lane information is an essential part of high-resolution micro-traffic data (HRMTD). Most of the lane detection algorithms for Light Detection and Ranging (LiDAR) are applied to high-density onboard LiDAR and airborne LiDAR, which cannot be used directly to process low-density roadside LiDAR data. In this article, an algorithm for lane detection applied to low-density roadside LiDAR is proposed, which includes three steps: ground recognition, lane marking point extraction, and pavement lane marking points clustering. Firstly, an improved algorithm for ground recognition is proposed to get normal ground points (NGPs). Then, the lane marking points are extracted from the no lane marking points based on the difference of laser intensity. Finally, we divided the ground within the range of LiDAR scanning into different stripes, and search the maximum linear density of the stripe to extract the lane markings. The average distance error (ADE) of lane detection results is less than 0.1m by fusing and comparing the test results with the aerial photographs. The proposed algorithm is also proved to be robust in multiple test locations and the results can be used to assist high-precision vehicle positioning in vehicle-to-infrastructure (V2I) cooperation application within the intelligent transportation system.

Automatic Vehicle Tracking with LiDAR-Enhanced Roadside Infrastructure

Abstract Vehicle tracking technology is a prerequisite for the connected-vehicle (CV) system. However, a mixture of CV and unconnected vehicles will be under normal conditions on roads in the near future. How to obtain the real-time traffic status of unconnected vehicles remains a challenge for traffic engineers. The roadside Light Detection and Ranging (LiDAR) sensor provides a solution for collecting real-time high-resolution micro traffic data of all road users (CV and unconnected vehicles). This article developed a systematic procedure for vehicle tracking using the roadside LiDAR sensors. The procedure can be divided into five major parts: point registration, background filtering, point clustering, object classification, and vehicle tracking. For each step, the corresponding data processing algorithms were provided. A field test was conducted to evaluate the performance of the proposed method. Compared to the state-of-the-art method, the proposed methods can track vehicles with higher accuracy and lower computation loads.