Clothoidal Mapping of Road Line Markings for Autonomous Driving High-Definition Maps

Gallazzi, Barbara; Cudrano, Paolo; Frosi, Matteo; Mentasti, Simone; Matteucci, Matteo

doi:10.1109/iv51971.2022.9827028

Cited by 7 publications

(4 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Visual localization method Visual map construction HD map [29][30][31][32][33] Large data volume, high storage requirements, and low matching efficiency…”

Section: Classical Localization In Unknown Environments Localization ...mentioning

confidence: 99%

“…An HD visual map can provide accurate and stable position references for localization [29], thereby minimizing cumulative errors. For example, Jingyu Li et al proposed a lane-segmentation detection method, named Lane-DeepLab, for detecting multi-class lane lines in unmanned driving scenarios for high-definition maps [30]; Huayou Wang et al extracted semantic information from an HD map and performed data association (DA) to obtain high-precision vehicle poses [31]; Gallazzi et al exploited a line-detection algorithm to retrieve the required information from camera images, thus constructing a lane-level HD map [32]; and Tuopu Wen et al introduced the use of an abundance of visual features and multi-frame HD map landmark features to constrain the estimation of the vehicle position and improve localization accuracy [33]. Although HD maps provide rich information support, they have disadvantages such as a large data volume, high storage requirements, and a low matching efficiency.…”

Section: Map-based Localizationmentioning

confidence: 99%

“…Table 2 provides a comparison of the memory requirements for the maps constructed by the proposed method and Tao's method [32]. The proposed method reduced map memory usage by over 75% compared with Tao's method.…”

Section: Maps Constructionmentioning

confidence: 99%

See 2 more Smart Citations

Enhanced Berth Mapping and Clothoid Trajectory Prediction Aided Intelligent Underground Localization

Li,

Chen,

Yuan

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

In response to the widespread absence of global navigation satellite system (GNSS) signals in underground parking scenes, we propose a multimodal localization method that integrates enhanced berth mapping with Clothoid trajectory prediction, enabling high-precision localization for intelligent vehicles in underground parking environments. This method began by constructing a lightweight map based on the key berths. The map consisted of a series of discrete nodes, each encompassing three elements: holistic and local scene features extracted from an around-view image, and the global pose of the mapping vehicle calculated using the positions of the key berth’s corner points. An adaptive localization strategy was employed during the localization phase based on the trajectory prediction result. A progressive localization strategy, relying on multi-scale feature matching, was applied to the nodes within the map coverage range. Additionally, a compensation localization strategy that combined odometry with the prior pose was utilized for the nodes outside the map coverage range. The experiments conducted in two typical underground parking scenes demonstrated that the proposed method achieved a trajectory prediction accuracy of 40 cm, a nearest map search accuracy exceeding 92%, and a metric localization accuracy meeting the 30 cm standard. These results indicate that the proposed approach satisfies the high-precision, robust, real-time localization requirements for intelligent vehicles in underground parking scenes, while effectively reducing the map memory requirements.

show abstract

“…Visual localization method Visual map construction HD map [29][30][31][32][33] Large data volume, high storage requirements, and low matching efficiency…”

Section: Classical Localization In Unknown Environments Localization ...mentioning

confidence: 99%

Section: Map-based Localizationmentioning

confidence: 99%

See 1 more Smart Citation

Enhanced Berth Mapping and Clothoid Trajectory Prediction Aided Intelligent Underground Localization

Li,

Chen,

Yuan

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…The road surface models established by the above methods are relatively simple, and they are difficult to fit to the real road surface well. For this reason, some references have introduced more complex road surface profile models, such as clothoid [23] and B-spline surface [24]. However, because the road surface usually presents many local features (such as local protrusions and depressions), it is difficult to use a unified model to describe it.…”

Section: Related Workmentioning

confidence: 99%

A Novel Method for Obstacle Detection in Front of Vehicles Based on the Local Spatial Features of Point Cloud

Lin

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Obstacle detection is the primary task of the Advanced Driving Assistance System (ADAS). However, it is very difficult to achieve accurate obstacle detection in complex traffic scenes. To this end, this paper proposes an obstacle detection method based on the local spatial features of point clouds. Firstly, the local spatial point cloud of a superpixel is obtained through stereo matching and the SLIC image segmentation algorithm. Then, the probability of the obstacle in the corresponding area is estimated from the spatial feature information of the local plane normal vector and the superpixel point-cloud height, respectively. Finally, the detection results of the two methods are input into the Bayesian framework in the form of probabilities for the final decision. In order to describe the traffic scene efficiently and accurately, the detection results are further transformed into a multi-layer stixel representation. We carried out experiments on the KITTI dataset and compared several obstacle detection methods. The experimental results indicate that the proposed method has advantages in terms of its Pixel-wise True Positive Rate (PTPR) and Pixel-wise False Positive Rate (PFPR), particularly in complex traffic scenes, such as uneven roads.

show abstract

TerrainSense: Vision-Driven Mapless Navigation for Unstructured Off-Road Environments

Hassan,

Sharma,

Madjid

et al. 2024

2024 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

Clothoidal Mapping of Road Line Markings for Autonomous Driving High-Definition Maps

Cited by 7 publications

References 32 publications

Enhanced Berth Mapping and Clothoid Trajectory Prediction Aided Intelligent Underground Localization

Enhanced Berth Mapping and Clothoid Trajectory Prediction Aided Intelligent Underground Localization

A Novel Method for Obstacle Detection in Front of Vehicles Based on the Local Spatial Features of Point Cloud

TerrainSense: Vision-Driven Mapless Navigation for Unstructured Off-Road Environments

Contact Info

Product

Resources

About