Deep Learning-based Intrusion Detection Approach for Autonomous Electric Vehicles

Technological advancements have given rise to Autonomous Electric Vehicles (AEVs), self-driving and independent computational entities integrated into the IoT landscape, promising enhanced efficiency in the transportation ecosystem. However, within this intricate web of interconnectedness of IoT network which used AEVs, ensuring the security of data transmission and actuation takes center stage as a critical and paramount concern. Preserving the integrity and confidentiality of data in this complex IoT environment becomes essential to thwart potential malicious attacks. To confront this pressing challenge, we propose a state-of-the-art approach centered around the utilization of Convolutional Neural Networks (CNNs). This novel approach is meticulously designed for real-time threat detection, culminating in a robust alert system that fortifies AEV security within the IoT ecosystem. Loss and Accuracy curves with confusion matrix is included in the paper to evaluate performance of CNN. Rigorous evaluation, including diverse optimizers and comprehensive metrics, underscores the approach's reliability, affirming its potential in safeguarding IoT-based autonomous transportation.