By Kyle Wiggers
Is it really possible to distinguish among cars, trucks, and pedestrians with radar alone? Absolutely, as it turns out, and all thanks to AI. In a newly published paper on the preprint server Arxiv.org (“Radar-based Road User Classification and Novelty Detection with Recurrent Neural Network Ensembles“), scientists at Daimler and the University of Kassel in Kassel, Germany describe a novel machine learning model that categorizes individual “traffic participants” — including hidden object classes which weren’t previously known to it — from radar data alone. They claim that their approach could be of particular usefulness to the driverless car industry, where object detection remains an acute area of interest.
The researchers sourced a data set containing more than 3 million data points on 3,800 instances of moving road users. Samples were acquired with four radar sensors mounted on the front half of a test vehicle, with a ranges of roughly 100 meters, and detected objects were slotted into one of six buckets: pedestrian, pedestrian group, bike, car, truck, and garbage. The label “pedestrian group” was attributed to multiple pedestrians which couldn’t be clearly separated in the data, while the “garbage” and “other” classes consisted of wrongly detected artifacts and road users which didn’t fit into any of the aforementioned groups, respectively (like motorcyclists, scooters, wheelchair users, cable cars, and dogs).
Image Credit: VentureBeat