Active Federated YOLOR Model for Enhancing Autonomous Vehicles Safety

3Citations
Citations of this article
7Readers
Mendeley users who have this article in their library.
Get full text

Abstract

A precise and real-time object detection system is crucial to ensuring the safety, smoothness, and trust of Autonomous Vehicles (AVs). Several machine learning techniques have been designed to improve vehicle detection capabilities and reduce the shortcomings caused by limited data and by transferring these data to a central server, which has shown poor performance under different conditions. In this paper, we propose an active federated learning-integrated solution over AVs that capitalizes on the You Only Learn One Representation (YOLOR) approach, a Convolutional Neural Network (CNN) specifically designed for real-time object detection. Our approach combines implicit and explicit knowledge, together with active learning and federated learning with the aim of improving the detection accuracy. Experiments show that our solution achieves better performance than traditional solutions (i.e., Gossip decentralized model and Centralized model).

Cite

CITATION STYLE

APA

Rjoub, G., Bentahar, J., & Joarder, Y. A. (2022). Active Federated YOLOR Model for Enhancing Autonomous Vehicles Safety. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 13475 LNCS, pp. 49–64). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-14391-5_4

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free