Modern vehicles increasingly rely on millimeter-wave (mmWave) radar sensors to support driver-assistance and automated-driving capabilities. However, each vehicle still perceives the world only from its own vantage point. Buildings, large trucks, adverse weather, clutter, and simple distance limits can therefore hide critical hazards, such as a pedestrian entering a crosswalk, a vehicle approaching a blind intersection, or a fast-sudden lane merging conflict. While connected-vehicle technologies allow vehicles to exchange messages, they do not currently enable vehicles to share radar-based understanding of the surrounding scene in a way that is timely, compact, and directly useful for safety‑critical decision making. This project explores a new paradigm in which the radar already installed on a vehicle becomes an active part of a wireless network. Nearby vehicles and roadside infrastructure cooperatively share information to construct a richer and more reliable view of the roadway than any single platform could form alone. This project will develop artificial‑intelligence methods that allow each vehicle to convert its radar measurements into compact summaries of the surrounding environment, determine which information is most important and urgent to share, and transmit that information efficiently over bandwidth‑limited and rapidly changing wireless links. The project will combine theory, large‑scale simulation, and laboratory‑scale testbed prototyping. It also includes outdoor experimentation on national wireless research platforms. Together, these efforts will test whether cooperative radar networking can extend perception beyond line of sight without requiring expensive new sensing hardware. If successful, this work could improve roadway safety and inform the design of future cellular vehicle‑to‑everything (C‑V2X) and 6G networks. The underlying ideas also apply to drone swarms, warehouse robotics, and smart‑city sensing systems. The project will train stu