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

2026-05-11

AI Summary

This work addresses the challenges of high communication overhead and susceptibility to interference from manhole covers and speed bumps in large-scale urban pothole detection. To this end, the authors propose a federated edge-cloud collaborative framework for temporal signal segmentation. On the vehicle side, a lightweight Gaussian Mixture Model performs initial anomaly screening with high recall, drastically reducing unnecessary data uploads. The cloud then applies a one-dimensional attention U-Net to conduct fine-grained temporal segmentation on the filtered signals, leveraging federated learning to aggregate knowledge from distributed vehicles. Experimental results demonstrate that the proposed approach not only lowers communication costs but also enhances discrimination against confounding road anomalies, achieving more accurate pothole boundary detection under both centralized and federated settings.

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.

Problem

• pothole detection
• vibration sensing
• communication overhead
• road event discrimination
• urban road networks

Innovation

• edge-cloud collaboration
• federated learning
• temporal segmentation
• vibration-based pothole detection
• onboard event screening