Graph Laplacian Diffusion Localization of Connected and Automated Vehicles

TL;DR

This paper introduces distributed graph-based localization methods for connected vehicles, leveraging diffusion strategies and adaptive algorithms to improve position accuracy and outperform existing solutions.

Contribution

It presents novel multi-modal localization approaches using graph diffusion and adaptive algorithms for connected vehicles, enhancing accuracy and robustness.

Findings

Significantly reduce GPS error in vehicle localization

Outperform state-of-the-art localization methods

Robust to network latency and connection variations

Abstract

In this paper, we design distributed multi-modal localization approaches for Connected and Automated vehicles. We utilize information diffusion on graphs formed by moving vehicles, based on Adapt-then-Combine strategies combined with the Least-Mean-Squares and the Conjugate Gradient algorithms. We treat the vehicular network as an undirected graph, where vehicles communicate with each other by means of Vehicle-to- Vehicle communication protocols. Connected vehicles perform cooperative fusion of different measurement modalities, including location and range measurements, in order to estimate both their positions and the positions of all other networked vehicles, by interacting only with their local neighborhood. The trajectories of vehicles were generated either by a well-known kinematic model, or by using the CARLA autonomous driving simulator. The various proposed distributed and diffusion localization schemes significantly reduce the GPS error and do not only converge to the global solution, but they even outperformed it. Extensive simulation studies highlight the benefits of the various approaches, outperforming the accuracy of the state of the art approaches. The impact of the network connections and the network latency are also investigated.