New ViL Platform Validates Cooperative Perception for Autonomous Driving

European safety regulation now allows a large share of automated-driving homologation evidence to be produced virtually, provided a validated physical-virtual facility generates it. To meet this need, a European research team built and deployed a hybrid Vehicle-in-the-Loop (ViL) platform that couples a real instrumented vehicle with a CARLA-based digital twin through a V2X message pipeline. The system streams ETSI-compliant CAM/CPM messages from the real car into the digital twin, where a GPU-accelerated Cooperative Perception (CP) module fuses them into a probabilistic occupancy grid during scenario runtime. The platform was demonstrated on a multi-vehicle double T-intersection scenario, testing CP workload across nominal, rain, and night conditions, as well as five localization-noise levels.

The results show that Cooperative Perception substantially widens field-of-view coverage and improves occupied-cell recall. Critically, beyond a moderate localization-noise threshold, positioning uncertainty—not weather—becomes the dominant error source. The team discussed the platform's current architectural limits and defined engineering targets to address them. The work, presented at IEEE ITSC 2026 Industry Track, outlines a trajectory toward a Mediterranean operational design domain (ODD) testing service, offering a practical path for validating autonomous driving systems in real-world-like conditions without full public-road exposure.