ORNL’s Real-Twin project is a streamlined scenario generation tool that automatically integrates real-world traffic data to create high-fidelity digital twins for simulating the impacts of connected and automated vehicles in microsimulation environments.
Physics-informed, RL-aligned evaluation engine for autonomous driving: JAX/Flax NNX trajectory-diffusion world models on the Waymo Open Dataset, DPO and reward-guided adversarial steering toward the scenarios that break AV stacks, physics feasibility, occupancy flow, WOSAC-style realism metrics and sensor simulation.
