Self-Supervised Learning for Enhancing Angular Resolution in Automotive MIMO Radars

A novel framework to enhance the angular resolution of automotive radars is proposed. An approach to enlarge the antenna aperture using artificial neural networks is developed using a self-supervised learning scheme. Data from a high angular resolution radar, i.e., a radar with a large antenna aperture, is used to train a deep neural network...

Dynamic Road Surface Signatures in Automotive Scenarios

A method to compute road surface signatures in the automotive scenario is presented. This method is subsequently applied to a rough asphalt surface without undulations. It is shown that, due to the observation geometry, the road surface signature experiences significant spreading along the Doppler-axis, resulting in a distinct range-Doppler map...