Review on Anomaly Detection in Autonomous Electric Vehicles Using Artificial Intelligence Technique

Engineering notes

Key topics: self-driving, control. See the paper for implementation details and experimental results.

Chinese explanation / 中文解读

中文解读待补充：本站会优先为端到端自动驾驶、BEV感知、3D目标检测、轨迹预测、路径规划、LiDAR感知等高价值论文补充中文说明。

Original abstract

The next wave in smart transportation is focused on the development of renewable energy sources that can help the automobile industry transition to self-driving electric automobiles (AEVs). In Internet-of-vehicles (IoV) ecosystems, AEVs are sensor-driven and driverless vehicles that leverage artificial intelligence (AI)-based interactions. AEVs can cut carbon emissions and trade energy with other AEVs, smart grids, and roadside units, among other things (RSUs). It is in favour of a more environmentally friendly mode of transportation. Sensor data, energy units, and user data, on the other hand, are shared through open networks, making them vulnerable to numerous security and privacy assaults. As a result, malevolent entities can remotely control and command AEVs, causing bogus updates to be propagated to peer nodes in an IoV context. This can result in component failure, congestion, and even the loss of data. Researchers and security experts throughout the world have addressed solutions that fulfil specific security requirements, but the detection and categorization of malicious AEVs is still a hot issue of research. Malicious AEVs have aberrant behavior that distinguishes them from regular AEVs, necessitating the identification of anomalous AEVs and the classification of anomaly types. The survey gives a systematic outlook on AI strategies in anomaly detection of AEVs, based on the aforementioned facts.

Links and sources

Comments