Real-Time Emergency Vehicle Detection and Response in AVS
Abstract
Autonomous vehicles changing urban traffic raises challenges for guaranteeing their and EVs’ safety and smooth operations in urgent situations. The SmartAV-RescueSense system is presented here as a multi-sensor and artificial intelligence framework made to help AVs notice and answer to emergency vehicles right away. With the help of its new fusion engine, the system combines microphone signals from different directions and all-around cameras to properly sense sirens and emergency signs, regardless of whether it’s noisy or hard to see. Because of the module, emergency signals are quickly recognized and the system functions well without the need for online storage or processing. Besides, the platform uses the Vehicle-to-Everything (V2X) protocol to obtain live warnings from equipped emergency vehicles, allowing AVs to act safely and efficiently. Thanks to its reinforcement learning structure, the dynamic path model directs AVs to give way or change their route in the best way given the situation and the urban environment. The approach also makes use of mapped priority regions over time to anticipate the route of an emergency vehicle in a busy city grid. Simulations of many different emergencies at the same time confirm that the system works well and responds promptly. It shows that this method represents considerable advantages in accuracy, time for response, and safety over older technologies. thanks to this research, urban traffic safety is increasing and these cars can live side by side smoothly. Those trying to improve these systems will increasingly focus on significant, real-life experiments and fitting them into city infrastructure.