Construction and Verification of a High-Precision Base Map for an Autonomous Vehicle Monitoring System

Ham, Sungil; Im, Jun-Hyuck; Kim, Minjun; Cho, Kuk

doi:10.3390/ijgi8110501

Cited by 8 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Forward moving vehicles maintain their speed, no matter how fast they are, but the speed of turning vehicles must adhere to geometric and surface characteristics to avoid skidding and run-off-way. The precision required in the AV movement has also been assumed in most of the references (Ham et al, 2019;Yuan et al, 2017). In contrast, previous literature were only trying to develop theories and rules for AVs movement at intersections without any improvement or consideration of the intersection capacity (Dresner and Stone, 2004;Milanés et al, 2010).…”

Section: Discussionmentioning

confidence: 99%

A Grid-based Framework for Managing Autonomous Vehicles' Movement at Intersections

Ghadi

2024

Period. Polytech. Transp. Eng.

View full text Add to dashboard Cite

With a full fleet of autonomous vehicles (AV) in the future, the concept of signal-free intersections has attracted transportation researchers. Multiple studies have investigated different methods and protocols to facilitate vehicle decision-making at intersections. Most of these methods have relied on strong centralized or inter-vehicle communications. This research aims to develop a new protocol for managing vehicle movement at four-leg intersections assuming a fully automated fleet. The main concept of the proposed protocol is to maintain a continuous movement of vehicles entering the intersection, without stopping in queues, by controlling their sequence of movements. In the methodology a dynamic occupancy grid (DOG) approach is applied by initially dividing the intersection into dynamic grids (i.e. cells). The cells are in a virtual movement, emanating away from each lane-group without overlapping, like a gear machine. Each vehicle sits on a specific cell to traverse the intersection safely at a predetermined time. In other words, vehicles approaching an intersection must register their speed and position by certain sensors, and in return receive the appropriate acceleration and speed to finally be allocated to the suitable moving grids. The efficiency of the applied protocol was demonstrated by a practical example that presented a higher intersection capacity value exceeding the ideal saturation flow rate in some cases. This reflects the efficiency of the implemented protocol and its applicability to different low to high traffic flows. Moreover, the protocol shows more flexibility in dealing with different weather, geometric and traffic conditions.

show abstract

Section: Discussionmentioning

confidence: 99%

A Grid-based Framework for Managing Autonomous Vehicles' Movement at Intersections

Ghadi

2024

Period. Polytech. Transp. Eng.

View full text Add to dashboard Cite

show abstract

“…False positives must be extremely low in order to be evaluated as a good model. Therefore, it is determined that the higher the recall value of 90% [1,44], the better the model. The change in the model performance is determined depending on the data combination.…”

Section: Discussionmentioning

confidence: 99%

“…A method for maintaining and rapidly updating changes in object data contained within the HD map is a major challenge. The equipment required to update HD maps usually involves an MMS equipped with light detection and ranging (LiDAR), a global navigation satellite system (GNSS), an inertial navigation system (INS), and vision sensors [1,2]. Unfortunately, equipment related to map renewal is very expensive, costing up to hundreds of millions of dollars.…”

Section: Introductionmentioning

confidence: 99%

Traffic Landmark Matching Framework for HD-Map Update: Dataset Training Case Study

Park

Woo³

et al. 2022

Electronics

View full text Add to dashboard Cite

High-definition (HD) maps determine the location of the vehicle under limited visibility based on the location information of safety signs detected by sensors. If a safety sign disappears or changes, incorrect information may be obtained. Thus, map data must be updated daily to prevent accidents. This study proposes a map update system (MUS) framework that maps objects detected by a road map detection system and the object present in the HD map. Based on traffic safety signs notified by the Korean National Police Agency, 151 types of objects, including traffic signs, traffic lights, and road markings, were annotated manually and semi-automatically. Approximately 3,000,000 annotations were trained based on the you only look once (YOLO) model, suitable for real-time detection by grouping safety signs with similar properties. The object coordinates were then extracted from the mobile mapping system point cloud, and the detection location accuracy was verified by comparing and evaluating the center point of the object detected in the MUS. The performance of the groups with and without specified properties was compared and their effectiveness was verified based on the dataset configuration. A model trained with a Korean road traffic dataset on our testbed achieved a group model of 95% mAP and no group model of 70.9% mAP.

show abstract

“…An HD map is created by processing data from a mobile mapping system equipped with LiDAR, GNSS, INS, and camera sensors [22], [23]. The mobile mapping system generates a geographic 3D point cloud by integrating local 3D point cloud data acquired from the LiDAR and geographic position and attitude of the system from the GNSS and INS.…”

Section: A High-definition Map and Reference Pointsmentioning

confidence: 99%

Semi-Automatic Framework for Traffic Landmark Annotation

Lee

Jung

Kang

et al. 2021

IEEE Open J. Intell. Transp. Syst.

View full text Add to dashboard Cite

We present a semi-automatic annotation method to build a large dataset for traffic landmark detection, where traffic landmarks include traffic signs, traffic lights as well as road markings. Labor-intensive bounding box tagging is a huge challenge to generate a large dataset for detection algorithms. To mitigate the labor, we adopt a high-definition (HD) map and a positioning system. We propose a process to align the HD map and images semi-automatically. Through the registration, the annotations of the HD map can be directly tagged onto traffic landmarks in the images. To make full use of the HD map for the dataset generation, we annotate the traffic landmarks with reference points, following the way that they are represented in the HD map, instead of the bounding boxes. The proposed semi-automatic method speeds up the annotation by a factor of 3.19, as compared to the manual annotation. Our dataset consists of about 150,000 images and includes about 470,000 annotated traffic landmarks. We train a deep neural network on our dataset to detect the traffic landmarks, and its performance is evaluated using a novel evaluation metric. Moreover, we show that the pretrained traffic landmark detection network is effective in detecting traffic landmarks in other countries using the bounding box by fine-tuning.

show abstract

Construction and Verification of a High-Precision Base Map for an Autonomous Vehicle Monitoring System

Cited by 8 publications

References 23 publications

A Grid-based Framework for Managing Autonomous Vehicles' Movement at Intersections

A Grid-based Framework for Managing Autonomous Vehicles' Movement at Intersections

Traffic Landmark Matching Framework for HD-Map Update: Dataset Training Case Study

Semi-Automatic Framework for Traffic Landmark Annotation

Contact Info

Product

Resources

About