Dynamic Deployment of Mobile Charging Trucks During Natural Disaster Evacuation: An Offline-to-Online Framework
Researchers have developed a new framework called ARMD to dynamically deploy mobile charging trucks (MCTs) for electric vehicles during natural disaster evacuations. This framework addresses the issue of overloaded fixed charging stations by coordinating multiple MCTs using a multi-agent proximal policy optimization approach. The system is trained offline and refined online, incorporating a spatio-temporal predictor for real-time route updates. Experiments in a simulated hurricane evacuation showed ARMD significantly reduces risk exposure compared to existing methods, particularly under disruptions like infrastructure failures. AI
IMPACT This framework could improve emergency response logistics by optimizing resource deployment for electric vehicles during crises.