Position and Velocity Fusion Using Multiple Monostatic Radar Sensors for Automotive Applications

In order to protect vulnerable road users (VRU), it is essential to detect and track them reliably and to predict their direction of motion i.e., their velocity vector relative to a vehicle's driving direction. A fusion approach for monostatic radar sensors, which computes target positions and the assigned two-dimensional instantaneous velocities, is proposed to realize these functionalities. These fused detections are used to cluster extended objects. Afterwards their position, velocity, and covariance matrices are estimated. A converted measurement Kalman filter is utilized to track these detected objects. Both velocity and position can be very accurately measured with the proposed fusion algorithm. Tracking by two separated radar stations greatly outperforms a single sensor setup, as verified by test data.