Cooperative Robotics Reinforced by Collective Perception for Traffic Moderation

Researchers have introduced a novel traffic safety system utilizing a cooperative humanoid robot to prevent collisions at non-line-of-sight (NLOS) intersections. Addressing the limitations of current V2X technology, which suffers from low vehicle penetration and driver inattention, this solution employs collective perception to extend awareness. The system integrates data from dual-camera infrastructure units detecting vehicle dynamics and onboard V2X units receiving cooperative awareness messages. A fusion module processes these inputs to maintain a real-time view of the main road, defining a Zone of Danger to predict collision risks for merging users. When a hazard is identified, the robot physically blocks the merging path and issues a human-like STOP gesture until the threat passes. Deployed and tested at the Future Mobility Park in Rotterdam, experiments demonstrated that combining vision and V2X perception enables early detection of approaching vehicles and reliable prevention of unsafe merges in real-world conditions. This approach effectively bridges the gap between connected and unconnected vehicles, offering a robust method for active traffic moderation in complex urban environments where visibility is limited.