Edge-Cloud Collaborative Pothole Detection via Onboard Event Screening and Federated Temporal Segmentation

TL;DR

This paper introduces an edge-cloud framework for pothole detection using vehicle vibration sensors, employing onboard screening and federated learning to reduce data transmission and improve detection accuracy.

Contribution

It proposes a novel combination of onboard GMM-based event screening and federated temporal segmentation with a 1D Attention U-Net for accurate pothole detection.

Findings

Reduces data transmission by filtering non-pothole events onboard.

Improves pothole detection accuracy with federated learning.

Effective in multi-vehicle experiments under various data distributions.

Abstract

Road potholes threaten driving safety and increase infrastructure maintenance costs, while large-scale and timely pothole detection remains challenging in urban road networks. Vehicle-mounted vibration sensing offers a low-cost and scalable solution, however, continuous transmission of raw acceleration streams causes high communication overhead. Also, vibration patterns induced by potholes are often confused with those caused by manholes, speed bumps, and other local road structures. To address these challenges, this paper proposes an edge-cloud collaborative pothole detection framework based on onboard vibration event screening and federated temporal segmentation. At the vehicle side, a Gaussian Mixture Model (GMM)-based module adaptively models background vibration and screens candidate abnormal events from continuous acceleration streams. The onboard module acts as a lightweight high-recall filter and uploads only compact candidate event segments with their contextual information. At the server side, pothole detection is formulated as a point-wise temporal segmentation task. A 1D Attention U-Net is developed to distinguish potholes from vibration-similar road events by capturing multi-scale temporal features and preserving event boundary information. Furthermore, the model is trained under a federated learning framework to exploit distributed multi-vehicle data while accommodating non-IID vehicle data distributions. Experiments on multi-vehicle vibration sensing data show that the proposed framework reduces unnecessary data transmission from smooth road segments and improves fine-grained pothole detection under both centralized and federated settings.