English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep learning (DL) has recently attracted increasing interest to improveobject type classification for automotive radar.In addition to high accuracy,it is crucial for decision making in autonomous vehicles to evaluate thereliability of the predictions; however, decisions of DL networks arenon-transparent. Current DL research has investigated how uncertainties ofpredictions can be quantified, and in this article, we evaluate the potentialof these methods for safe, automotive radar perception. In particular weevaluate how uncertainty quantification can support radar perception under (1)domain shift, (2) corruptions of input signals, and (3) in the presence ofunknown objects. We find that in agreement with phenomena observed in theliterature,deep radar classifiers are overly confident, even in their wrongpredictions. This raises concerns about the use of the confidence values fordecision making under uncertainty, as the model fails to notify when it cannothandle an unknown situation. Accurate confidence values would allow optimalintegration of multiple information sources, e.g. via sensor fusion. We showthat by applying state-of-the-art post-hoc uncertainty calibration, the qualityof confidence measures can be significantly improved,thereby partiallyresolving the over-confidence problem. Our investigation shows that furtherresearch into training and calibrating DL networks is necessary and offersgreat potential for safe automotive object classification with radar sensors.

Investigation of Uncertainty of Deep Learning-based Object Classification on Radar Spectra