Spinning radar Doppler velocity boosts vehicle detection and tracking accuracy
Do Spinning Radar Doppler Velocity Measurements Improve Vehicle Detection and Tracking?
RoboticsComputer Vision and Pattern Recognition
Summary
Detecting and tracking vehicles using radar is important for self-driving cars. The authors studied whether measuring how fast parts of a scene are moving (Doppler velocity) using spinning radars helps this process. They created a large labeled dataset by combining lidar and radar data to train their system. Their experiments show that correcting radar images with Doppler information improves detection slightly, while using Doppler velocities greatly improves how well moving vehicles are tracked over time.
What this means in practice
- •For autonomous vehicle developers: Improve vehicle detection and tracking systems by integrating Doppler velocity corrections from spinning radars for better perception in self-driving cars.$Commercial implications: Enables automotive companies to enhance radar-based perception modules, improving vehicle safety features and autonomy.
- •For traffic monitoring operators: Use enhanced radar Doppler velocity data to track vehicle movements more accurately in traffic surveillance setups.
Authors
Eric Xie, Daniil Lisus, Timothy D. Barfoot
Abstract
Spinning frequency-modulated continuous-wave (FMCW) radars have been gaining popularity in autonomous vehicle perception on account of their robustness to adverse weather conditions and 360° field of view. Recently, scanning radars have also been shown capable of generating per-azimuth Doppler velocity. In this paper, we investigate whether these Doppler velocity measurements improve spinning radar vehicle detection and tracking performance. For detection, we estimate the ego motion and use it to undo the Doppler range distortion of the radar image before passing it to a network. For tracking, we propose a new way to estimate a per-vehicle velocity and use it as a prior for the tracker's motion model. Since Doppler-enabled spinning radar data is not available in any dataset with ground-truth dynamic object labels, our first contribution is an automatic labelling pipeline that uses an ensemble of fine-tuned off-the-shelf lidar detectors to label all 643 km of the Boreas Road Trip dataset. We then transfer detections to radar, and use over 250 km of vehicle-dense sequences as ground-truth training data. By training and evaluating two state-of-the-art detectors, we show that Doppler undistortion can improve detection accuracy by up to $2.37$ points on mean average precision. Furthermore, we show that the Doppler velocity prior can improve tracking accuracy by $13.68$ points on multi-object tracking accuracy (MOTA) versus the zero-velocity initialization baseline, while achieving $99.7\%$ of the MOTA obtained using ground-truth velocities as the prior.