Automated Vehicle Radar Yaw-Rate Estimation

The patent segments the extended target into multiple scattering-points, each treated as an independent tracking entity. The controller tracks each scattering-point individually through successive radar scans, maintaining separate range, range-rate, and azimuth measurements for each point. This segmentation allows the system to handle the complexity of extended targets while using a single radar sensor, resolving the contradiction between device simplicity and measurement precision.

The patent implements a feedback mechanism where the controller continuously updates the state of each scattering-point based on new radar measurements and previous state information. By recursively processing range, range-rate, and azimuth data from multiple scattering-points across successive scans, the system refines its estimation of the target's yaw-rate and over-the-ground velocity, improving measurement precision without requiring multiple sensors.

The patent performs preliminary processing of raw radar detections by organizing range, range-rate, and azimuth data from multiple scattering-points before computing yaw-rate and over-the-ground velocity. This preliminary organization and filtering of data reduces noise and variability, enabling accurate real-time estimation without sacrificing measurement precision.

The patent transforms raw radar parameters (range, range-rate, azimuth) into derived parameters (yaw-rate, over-the-ground velocity) through mathematical relationships. By changing the parameter representation and using recursive estimation techniques, the system achieves both real-time productivity and improved measurement precision despite the inherent noise in raw radar data.