CADET: A Modular Platform for Evaluating Distributed Cooperative Autonomy in Connected Autonomous Vehicles
Researchers have developed CADET, a modular platform designed to evaluate distributed cooperative autonomy in connected autonomous vehicles. This system addresses the complexities of integrating deep learning models across vehicles, roadside units, and cloud infrastructure, accounting for network latency and compute heterogeneity. Experiments demonstrate that vehicle-to-vehicle communication for intent packets offers superior safety compared to cloud-based perception, and roadside unit-assisted perception remains robust until overloaded. AI
IMPACT Provides a framework for researchers to benchmark distributed inference workloads for autonomous vehicle systems.