Beyond Point Clouds: A Knowledge-Aided High Resolution Imaging Radar Deep Detector for Autonomous Driving

TL;DR

This paper introduces a knowledge-aided high-resolution radar imaging method combined with deep learning for improved object detection in autonomous driving, demonstrating high accuracy in real-world experiments.

Contribution

It proposes a novel MIMO radar processing chain with waveform orthogonality and a deep neural network for high-resolution object detection in autonomous driving.

Findings

Achieved 96.1% average precision in object detection

Developed a waveform orthogonality scheme for multiple transmit antennas

Validated the approach with field experiments using multi-modal sensors

Abstract

The potentials of automotive radar for autonomous driving have not been fully exploited. We present a multi-input multi-output (MIMO) radar transmit and receive signal processing chain, a knowledge-aided approach exploiting the radar domain knowledge and signal structure, to generate high resolution radar range-azimuth spectra for object detection and classification using deep neural networks. To achieve waveform orthogonality among a large number of transmit antennas cascaded by four automotive radar transceivers, we propose a staggered time division multiplexing (TDM) scheme and velocity unfolding algorithm using both Chinese remainder theorem and overlapped array. Field experiments with multi-modal sensors were conducted at The University of Alabama. High resolution radar spectra were obtained and labeled using the camera and LiDAR recordings. Initial experiments show promising performance of object detection using an image-oriented deep neural network with an average precision of 96.1% at an intersection of union (IoU) of typically 0.5 on 2,000 radar frames.