A Safety-Adapted Loss for Pedestrian Detection in Autonomous Driving

Maria Lyssenko, Piyush Pimplikar, Maarten Bieshaar, Farzad Nozarian, Rudolph Triebel

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In safety-critical domains like autonomous driving (AD), errors by the object detector may endanger pedestrians and other vulnerable road users (VRU). As raw evaluation metrics are not an adequate safety indicator, recent works leverage domain knowledge to identify safety-relevant VRU, and to back-annotate the criticality of the interaction to the object detector. However, those approaches do not consider the safety factor in the deep neural network (DNN) training process. Thus, state-of-the-art DNN penalize all misdetections equally irrespective of their importance for the safe driving task. Hence, to mitigate the occurrence of safety-critical failure cases like false negatives, a safety-aware training strategy is needed to enhance the detection performance for critical pedestrians. In this paper, we propose a novel, safety-adapted loss variation that leverages the estimated per-pedestrian criticality during training. Therefore, we exploit the reachable set-based time-to-collision (TTCRSB) metric from the motion domain along with distance information to account for the worst-case threat. Our evaluation results using RetinaNet and FCOS on the nuScenes dataset demonstrate that training the models with our safety-adapted loss function mitigates the misdetection of safety-critical pedestrians with robust performance for the general case, i.e., safety-irrelevant pedestrians.

Original languageEnglish
Title of host publication2024 IEEE International Conference on Robotics and Automation, ICRA 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages4428-4434
Number of pages7
ISBN (Electronic)9798350384574
DOIs
StatePublished - 2024
Externally publishedYes
Event2024 IEEE International Conference on Robotics and Automation, ICRA 2024 - Yokohama, Japan
Duration: 13 May 202417 May 2024

Publication series

NameProceedings - IEEE International Conference on Robotics and Automation
ISSN (Print)1050-4729

Conference

Conference2024 IEEE International Conference on Robotics and Automation, ICRA 2024
Country/TerritoryJapan
CityYokohama
Period13/05/2417/05/24

Fingerprint

Dive into the research topics of 'A Safety-Adapted Loss for Pedestrian Detection in Autonomous Driving'. Together they form a unique fingerprint.

Cite this