Research papers · Posted 2026.10.26 · Preprint
Uncertainty Evaluation to Support Safety of the Intended Functionality Analysis for Determining the Applicability of Performance Targets Across Triggering Conditions in ML-Based LiDAR Object Detection
Uncertainty Evaluation to Support Safety of the Intended Functionality Analysis for Determining the Applicability of Performance Targets Across Triggering Conditions in ML-Based LiDAR Object Detection
Overview
The study tests whether uncertainty thresholds selected in one condition preserve detection performance targets in another. Its LiDAR experiments emphasize frame dependence, independent sequence counts and retained correct detections when interpreting SOTIF evidence.
Publication and version information
This expanded manuscript remains a preprint. The author-reported acceptance or publication relates to a shorter earlier version; source wording is retained below.
Author-supplied source information
Author-supplied information on arXiv.
24 pages, 6 figures. Extended version of the VEHITS 2026 conference paper (doi:10.5220/0014967200004030). Submitted for journal publication. Code and recorded results: https://doi.org/10.5281/zenodo.23495356
Version: v1