In the realm of energy and transportation, a significant shift is underway with the advent of Connected and Autonomous Electrified Vehicles (CAEVs). These vehicles promise not only a cleaner environment but also more efficient traffic flow. However, their autonomous and connected nature makes them vulnerable to cyber-attacks. Researchers Shashank Dhananjay Vyas and Satadru Dey from the University of California, Berkeley, have been exploring ways to mitigate these risks.
Vyas and Dey have proposed a secure control architecture specifically designed for CAEVs. Their approach involves the creation of an additional control input using Reinforcement Learning (RL). This input is applied to the vehicle’s powertrain, complementing the input commanded by the battery. The goal is to ensure the safe operation of CAEV platoons, even in the face of adversarial attacks, by minimizing their impact.
The researchers demonstrated the potential of their approach through simulation case studies. These studies showed that the proposed method could effectively prevent collisions within CAEV platoons, even when under attack. This is a crucial step towards ensuring the safety and reliability of CAEVs, which are poised to play a significant role in the future of smart mobility.
The research was published in the IEEE Transactions on Intelligent Transportation Systems, a reputable journal in the field of transportation and vehicle technology. The findings could have practical applications for the energy sector, particularly in the development of smart grids and the integration of electric vehicles into these systems. By ensuring the secure and safe operation of CAEVs, the energy sector can better plan for the future of transportation and its impact on energy demand and distribution.
This article is based on research available at arXiv.