Anomaly Detection in Radar Data Using PointNets

TL;DR

This paper introduces a novel PointNet-based method for detecting anomalous radar targets in autonomous driving, addressing challenges like ghost targets and clutter with promising real-world results.

Contribution

It develops a new grouping variant within PointNet architecture tailored for radar anomaly detection, enhancing robustness against clutter and ghost targets.

Findings

Effective detection of anomalous radar targets in urban scenarios

Improved robustness over traditional methods

Promising results on real-world datasets

Abstract

For autonomous driving, radar is an important sensor type. On the one hand, radar offers a direct measurement of the radial velocity of targets in the environment. On the other hand, in literature, radar sensors are known for their robustness against several kinds of adverse weather conditions. However, on the downside, radar is susceptible to ghost targets or clutter which can be caused by several different causes, e.g., reflective surfaces in the environment. Ghost targets, for instance, can result in erroneous object detections. To this end, it is desirable to identify anomalous targets as early as possible in radar data. In this work, we present an approach based on PointNets to detect anomalous radar targets. Modifying the PointNet-architecture driven by our task, we developed a novel grouping variant which contributes to a multi-form grouping module. Our method is evaluated on a real-world dataset in urban scenarios and shows promising results for the detection of anomalous radar targets.