Federated Learning for Object Detection: Enabling Collaborative Drone Learning Without Centralizing Data

2026-07-02

AI Summary

This study addresses the challenges of data privacy, regulatory constraints, and communication bottlenecks inherent in centralized training for distributed unmanned aerial vehicle (UAV) systems by proposing a federated learning–based decentralized object detection approach. The method effectively applies federated learning to UAV edge scenarios for the first time, enabling collaborative training of a lightweight YOLOv5-nano model without sharing local image data. Experiments conducted on the KITTI-MiTA dataset using the Sherpa.ai platform demonstrate that, compared to standalone training, the proposed approach improves mAP@0.50 and mAP@0.50:0.95 by 52.89% and 67.80%, respectively, achieving performance close to that of centralized training. These results validate the feasibility of the method in enabling efficient, scalable, and privacy-preserving collaborative learning in distributed UAV environments.

Abstract

Object detection is a fundamental capability for AI-driven perception in safety-critical drone and edge-vision systems, including disaster response, operational security environments, infrastructure monitoring and defense applications. Robust model performance in such environments depends on large, continuously updated datasets. However, training high-performing detectors typically requires centralizing aerial imagery, which raises privacy, regulatory, storage, and bandwidth challenges. This is especially problematic in distributed drone deployments, where visual data is generated onboard and is often impractical or undesirable to transfer to a centralized infrastructure. In this work, we apply Federated Learning (FL) for object detection, enabling drones to improve a shared model while keeping image data local and private. We implement a federated object detection pipeline using the Sherpa.ai FL platform on the KIIT-MiTA dataset, and compare it with Single-drone and Centralized baselines using mean Average Precision (mAP) at IoU thresholds of 0.50 and 0.50-0.95. In our experiments, the proposed FL approach remains close to Centralized training while dramatically improving over Single-drone training, with the best lightweight model (YOLO26 nano), suitable for deployment even on very limited edge infrastructure, achieving relative gains of 52.89% and 67.80% in mAP@0.50 and mAP@0.50:0.95, respectively. These results show that FL enables scalable, high-performing, and privacy-preserving object detection across distributed drone fleets without data centralization.

Problem

• Federated Learning
• Object Detection
• Drone
• Data Privacy
• Edge Computing

Innovation

• Federated Learning
• Object Detection
• Drone
• Edge AI
• Privacy-Preserving