MDrive: Benchmarking Closed-Loop Cooperative Driving for End-to-End Multi-agent Systems
Marco Coscoy, Zewei Zhou, Seth Z. Zhao, Henry Wei, Angela Magtoto, Johnson Liu, Rui Song, Walter Zimmer, Zhiyu Huang, Chen Tang, Bolei Zhou, Jiaqi Ma
THE PROBLEM
This paper focuses on Control & PlanningPlanningFiguring out what the robot should do before or during movement.. MDrive provides the first closed-loop Simulation & Sim-to-RealBenchmarkA standard test used to compare methods fairly. for evaluating multi-agent autonomous driving with V2X communication, revealing that Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world. sharing doesn't guarantee better Control & PlanningPlanningFiguring out what the robot should do before or during movement. and negotiation can backfire in dense traffic. Developers can now test cooperative driving systems on 225 realistic scenarios with ground truth from real-world datasets and NHTSA typologies. Read the paper by tracking the Core ConceptsTaskThe job the robot is supposed to complete, such as pick-and-place, navigation, or drawer opening. definition, the Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. or data assumptions, and the evidence that supports the claimed improvement.
HOW IT WORKS
Task framing
Core method
Data and supervision
Evaluation evidence
KEY RESULTS
MDrive provides the first closed-loop Simulation & Sim-to-RealBenchmarkA standard test used to compare methods fairly. for evaluating multi-agent autonomous driving with V2X communication, revealing that Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world. sharing doesn't guarantee better Control & PlanningPlanningFiguring out what the robot should do before or during movement. and negotiation can backfire in dense traffic. Developers can now test cooperative driving systems on 225 realistic scenarios with ground truth from real-world datasets and NHTSA typologies.
WHY DEVELOPERS SHOULD CARE
MDrive provides the first closed-loop Simulation & Sim-to-RealBenchmarkA standard test used to compare methods fairly. for evaluating multi-agent autonomous driving with V2X communication, revealing that Perception & SensingPerceptionThe process of turning raw sensor data into useful understanding of the world. sharing doesn't guarantee better Control & PlanningPlanningFiguring out what the robot should do before or during movement. and negotiation can backfire in dense traffic. Developers can now test cooperative driving systems on 225 realistic scenarios with ground truth from real-world datasets and NHTSA typologies.
LIMITATIONS
The main limitation to check is whether the claimed behavior holds outside the paper's reported setup. That means testing across different Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. embodiments, scenes, objects, and data distributions.
WHAT COMES NEXT
The practical next step is independent reproduction with clear baselines, ablations, and stress tests. For a developer, the useful follow-up is to map the paper's Control & PlanningPlanningFiguring out what the robot should do before or during movement. assumptions onto a concrete Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. stack, then test the smallest version of the method that could run end to end.