Experimental Results of Target Classification Using mm Wave Corner Radar Sensors

Target classification based on the power intensity models has been used to post-process the measured data captured by on-vehicle mmWave radar sensors. By grouping the point targets with compensated ego motion, moving vehicles and pedestrians are identified. The measured intensity inclusive of all practical environmental effects has been used for calibration, and the relation between the intensity and the range are found based on a few scenario for demonstration purpose. As a reference, the model has been tested in collected data, the target capture rate is 65% while the classification accuracy reaches 90%. Further improvement can be expected when tracking model is included.