Artificial intelligence (AI) empowered anomaly detection for autonomous vehicles in 6G-V2X

Irfan Ali Kandhro; Fayyaz Ali; Ali Orangzeb Panhwar; Raja Sohail Ahmed Larik; Kanwal Fatima

doi:10.22581/muet1982.2303.09

Author(s)

Irfan Ali Kandhro Department of Computer Science, Sindh Madressatul Islam University,Karachi Sindh,Pakistan

Fayyaz Ali Department of Software Engineering, Sir Syed University of Engineering and Technology,Karachi Sindh,Pakistan

Ali Orangzeb Panhwar Department of Computer Science, Shaheed Zulfikar Ali Bhutto Institute of Science and Technology Gharo Sindh Pakistan,,

Raja Sohail Ahmed Larik School of Computer Science and Engineering, Nanjing University of Science and Technology,Nanjing, 210094,P.R. China

Kanwal Fatima Department of Computer Science, Sindh Madressatul Islam University, Karachi Sindh Pakistan

Abstract

The development of advanced Intelligent Transportation Systems has been made possible by the rapid expansion of autonomous vehicles (AVs) and networking technology (ITS). The in-vehicle users' increased data needs from AVs put the vehicle's trajectory data in danger and make it more susceptible to security threats. In this paper, Autonomous vehicles (AVs) transform the intelligent transportation system by exchanging real-time and seamless data with other AVs and the network (ITS). Transportation that is automated has many advantages for people. However, worries about safety, security, and privacy continue to grow. The AVs need to exchange sensory data with other AVs and with their own for navigation and trajectory planning. When an unreliable sensor-equipped AV or one that is malicious enters connectivity in such circumstances, the results could be disruptive. To effectively detect anomalies and mitigate cyberattacks in AVs, this study suggests the Efficient Anomaly Detection (EAD) method. The EAD technique finds and isolates rogue AVs using the Multi-Agent Reinforcement Learning (MARL) algorithm, which operates over the 6G network to thwart modern cyberattacks and provide a quick and accurate anomaly detection mechanism. The expected outcomes demonstrate the value of EAD and have an accuracy rate that is 8.01% greater than that of the current systems.

Publication Details

Page(s) 79-88

DOI 10.22581/muet1982.2303.09

Published Journal: Mehran University Research Journal of Engineering and Technology, Volume: 42, Issue: 3, Year: 2023

Request the full text from PASTIC

Keywords

Artificial Intelligence (AI) 6G-V2X Vehicle-to-everything (V2X) Intelligent reflective surfaces Autonomous Vehicles Multi-Agent Reinforcement Learning Maximum Entropy Inverse Reinforcement Learning