Real-Time V2V Communication for Collision Avoidance in AV Networks

The paper introduces a new Adaptive Decentralised Real-Time Vehicle-to-Vehicle (V2V) Communication Framework designed to enhance the efficiency of collision avoidance in autonomous vehicle (AV) networks. When AVs are found on more roads, fast and dependable communication between the vehicles is vital to avoid accidents and increase traffic safety. The framework introduces edge computing inside the vehicles, enabling them to process the data gathered by several sensors on the spot, thus cutting latency. Thanks to a blockchain-inspired decentralized protocol, all messages are authenticated and remain reliable without requiring central servers, thus increasing security. An AI-based collision forecasting system checks the collected sensor signals to predict collisions fast and take quick preemptive actions. It prioritizes the delivery of important messages and manages the device’s bandwidth by sending the signals only when needed. It also helps by switching available channels cleverly to prevent communication interruptions and overcrowding. Working together, models updates make the network better, and all data is kept confidential during this process. The results of the simulation explain that collision rates, time between messages, and network delays are reduced more than in current V2V systems. This approach makes it easy to avoid collisions in real time and offers good security, which also makes it simple for AVs to communicate with other vehicles on the roads. The suggested method is expected to make roads safer, increase traffic flow smoothness, and speed up the adoption of trustworthy autonomous vehicles.

Перевод пока недоступен