Automated Quantification of Occupant Posture and Shoulder Belt Fit Using Safety Specific Key Points

Franz Hartleitner,K Bohman,A Koppisetty

doi:10.1109/ojits.2022.3140612

Franz Hartleitner, K Bohman + Show 1 more

Open Access

https://doi.org/10.1109/ojits.2022.3140612

Copy DOI

Abstract

Virtual evaluation of automotive safety with variation in occupant posture and shoulder belt fit is gaining importance, and there is a need of methods facilitating analysis of occupant postures in driving studies. This study is aimed to develop an AI-based computer vision method to automatically quantify occupant posture and shoulder belt position over time in a car. Traceable defined key points on the occupant were related with the shoulder belt and quantified over time in real 3D coordinates by predefined key measurements, utilising the underlying spatial information of a Intel RealSense 3D Camera. The key points are defined as traceable key points relevant to relate the occupant to the vehicle environment and to estimate shoulder belt position. Key point prediction results suggest an average deviation of around 1cm per coordinate, which enable a reliable spatial categorization of the respective tracked occupant by analyzing the key measurements. This method providing continuous information of the occupant position and belt fit will be useful to identify common occupant postures as well as more extreme postures, to be used for expanding variations in postures for vehicle safety assessments.

Highlights

Occupant safety in vehicles is evaluated through crash tests, using anthropometric test devices (ATDs), which represent humans
The 2D result of the training process is reported in Table I in the metric Intersection over Union (IoU)
A compareable prediction quality is depicted in the Appendix in Figure 13, which shows a prediction of the test set with an IoU of 0.7

Summary

Introduction

Occupant safety in vehicles is evaluated through crash tests, using anthropometric test devices (ATDs), which represent humans. Legal requirements and consumer rating crash safety programs are well described in test protocols. The ATDs are positioned in standardized sitting postures [1]. In real life, the variation of sitting postures may be larger than those represented by the standardized sitting postures used in current crash tests [2]. Consumer rating programs like EuroNCAP [3] and IIHS [4] have started to explore virtual testing besides traditional crash tests and that opens up for parameter studies including a greater range of sitting posture. There is a need to increase the knowledge of variation in sitting postures that takes place in real life. Reliable data collected in real life conditions is needed to verify the existing protocols or show potential for improvement

Objectives

Methods

Results

Discussion

Conclusion