A Radar Target Tracking Algorithm Based on Learning Displacement

This letter considers the problem of target detection and tracking with millimeter-wave (mm-wave) frequency-modulated continuous-wave (FMCW) radar and proposes a tracking method via learning the displacement of the target in successive frames. First, a convolutional neural network (CNN), namely, detection and displacement network (DDNet), is trained to predict the target positions at current frame and the displacement relative to last frame simultaneously with the range-azimuth (RA) spectrum in two adjacent frames. Then, the detections are associated through the predicted displacement to construct the trajectories. Finally, the effectiveness of the proposed method is validated on both simulated and real-world datasets.