Risk Ranked Recall: Collision Safety Metric for Object Detection Systems in Autonomous Vehicles

Ayoosh Bansal, Jayati Singh, Micaela Verucchi, Marco Caccamo, Lui Sha

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

15 Scopus citations

Abstract

Commonly used metrics for evaluation of object detection systems (precision, recall, mAP) do not give complete information about their suitability of use in safety critical tasks, like obstacle detection for collision avoidance in Autonomous Vehicles (AV). This work introduces the Risk Ranked Recall (R3) metrics for object detection systems. The R3 metrics categorize objects within three ranks. Ranks are assigned based on an objective cyber-physical model for the risk of collision. Recall is measured for each rank.

Original languageEnglish
Title of host publication2021 10th Mediterranean Conference on Embedded Computing, MECO 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9780738133614
DOIs
StatePublished - 7 Jun 2021
Event10th Mediterranean Conference on Embedded Computing, MECO 2021 - Budva, Montenegro
Duration: 7 Jun 202110 Jun 2021

Publication series

Name2021 10th Mediterranean Conference on Embedded Computing, MECO 2021

Conference

Conference10th Mediterranean Conference on Embedded Computing, MECO 2021
Country/TerritoryMontenegro
CityBudva
Period7/06/2110/06/21

Keywords

  • Autonomous CPS
  • Dependable CPS
  • Safety

Fingerprint

Dive into the research topics of 'Risk Ranked Recall: Collision Safety Metric for Object Detection Systems in Autonomous Vehicles'. Together they form a unique fingerprint.

Cite this