Tsukada Lab. Computer Network Group

Cooperative Localization for Connected Autonomous Vehicles Utilizing LiDAR-Equipped Infrastructure

Yuze Jiang

, Mar 30, 2024

Summary in three lines

  • LiDAR map matching can become unreliable where detailed maps or distinctive landmarks are missing.
  • The thesis uses roadside LiDAR to observe vehicles and send additional localization information over V2I.
  • It evaluates how the infrastructure-assisted approach supplements vehicle-based localization.

Figures from the paper

Figure 4.1: Semantic diagram of the proposed cooperative localization system
Figure 4.1. Semantic diagram of the proposed cooperative localization systemSource: Figure 4.1 from Master's thesis, The University of Tokyo, 2024 (author's version)Vehicle information from CAM messages (size, heading, reference position) is sent to the roadside LiDAR, which refines the detected vehicle's bounding box and position and returns the localization result to the vehicle.
Figure 5.10: Localization error of roadside-LiDAR-based localization and NDT in Scenario 2
Figure 5.10. Localization error of roadside-LiDAR-based localization and NDT in Scenario 2Source: Figure 5.10 from Master's thesis, The University of Tokyo, 2024 (author's version)In the 0 to 30 m range of Scenario 2, roadside-LiDAR localization with the VLP-16 shows a lower median error than NDT, and both roadside LiDAR models vary less than NDT.

Abstract

Advancements in LiDAR technology have led to more cost-effective production while simultaneously improving precision and resolution. As a result, LiDAR has become in- tegral to vehicle localization, achieving centimeter-level accuracy through techniques like Normal Distributions Transform (NDT) and other advanced 3D registration algorithms. Nonetheless, these approaches are reliant on high-definition 3D point cloud maps, the cre- ation of which involves significant expenditure. When such maps are unavailable or lack sufficient features for 3D registration algorithms, localization accuracy diminishes, posing a risk to road safety. To address this, we proposed to use LiDAR-equipped roadside unit and Vehicle-to-Infrastructure (V2I) communication to accurately estimate the connected autonomous vehicle’s position and help the vehicle when its self-localization is not accu- rate enough. Our simulation results indicate that this method outperforms traditional NDT scan matching-based approaches in terms of localization accuracy.

How to cite

Y. Jiang, “Cooperative Localization for Connected Autonomous Vehicles Utilizing LiDAR-Equipped Infrastructure,” in , 2024.

BibTeX

@mastersthesis{Jiang2024,
  title         = {Cooperative Localization for Connected Autonomous Vehicles Utilizing LiDAR-Equipped Infrastructure},
  author        = {Yuze Jiang},
  year          = {2024},
  date          = {2024-03-30},
  urldate       = {2024-03-31},
  abstract      = {Advancements in LiDAR technology have led to more cost-effective production while simultaneously improving precision and resolution. As a result, LiDAR has become in- tegral to vehicle localization, achieving centimeter-level accuracy through techniques like Normal Distributions Transform (NDT) and other advanced 3D registration algorithms. Nonetheless, these approaches are reliant on high-definition 3D point cloud maps, the cre- ation of which involves significant expenditure. When such maps are unavailable or lack sufficient features for 3D registration algorithms, localization accuracy diminishes, posing a risk to road safety. To address this, we proposed to use LiDAR-equipped roadside unit and Vehicle-to-Infrastructure (V2I) communication to accurately estimate the connected autonomous vehicle’s position and help the vehicle when its self-localization is not accu- rate enough. Our simulation results indicate that this method outperforms traditional NDT scan matching-based approaches in terms of localization accuracy.},
  keywords      = {ITS, Thesis},
  pubstate      = {published},
  tppubtype     = {mastersthesis}
}