World Journal of Engineering Research and Technology

Mohammed Sharfuddin*

ABSTRACT

High-definition (HD) maps are essential for safe and efficient autonomous vehicle (AV) navigation. However, centralized HD map updating frameworks face limitations such as high latency, single points of failure, and user privacy concerns. In this paper, we propose a federated HD map updating framework that enables real-time, privacy- preserving collaboration between autonomous vehicles. Each vehicle detects road environment changes using AI-powered feature extraction and securely shares minimal metadata with a central map server using differential privacy and homomorphic encryption techniques. This federated approach ensures faster updates, distributed intelligence, and robust protection of sensitive location data. Our framework builds prior work in real-time HD map monitoring, AI-based change detection, and V2X security, and demonstrates improved map accuracy and update efficiency across a simulated AV fleet.

[Full Text Article] [Download Certificate]