Biomecanicamente.org actualidad on-line del Instituto de Biomecánica ■ Biomecanicamente.org IBV on-line news ■ Biomecánica, ciencia que estudia el comportamiento del cuerpo humano y su relación con los productos con que interactúa y el entorno en el que se desenvuelve ■ Biomechanics, the study of the human body's behaviour and its relation to both its surrounding, environment and the products it interacts with ■ Éxito empresarial a través del bienestar de las personas ■ Corporate success through people's wellbeing ■ Cuidamos tu calidad de vida ■ We take care of your quality of life ■ Automoción y medios de transporte ■ Automotive and mass transport sector ■
Wednesday, 10 May 2017 16:06

Using digital human models to study the ergonomics of the seated posture in the automotive sector Featured

Rate this item
(1 Vote)

Beatriz Nácher Fernández (1); Sandra Alemany Mut;(1) ; Fusako Sato(2); Jordi Uriel Molto(1); José S. Solaz Sanahuja(1); Elisa Signes i Pérez(1)

(1) Instituto de Biomecánica (IBV). Universitat Politècnica de València. Edificio 9C. Camino de Vera, s/n (46022) Valencia. Spain

(2) Japanese Automotive Research Institute

Using Digital Human Models in automotive sector is very popular, especially for impact simulation or to determine reaches and spaces for the driver and the passengers.           
Instituto de Biomecánica (IBV) has come up with the initial approach to develope three dimensional digital human models in the driving posture in the framework of a larger project sponsored by the JARI laboratory (Japanese Automotive Research Institute) of Japan, whose objective is to carry out research into the mechanics of whiplash in women, by developing, validating, and using finite element models (FEM) of the human body.                
The main contribution of the project has been the development of a methodology to obtain complete and simplified meshes of the human body in the driving posture. It has also been developed a procedure to scan people in the driving posture using a scanner that is specifically configured for the standing posture. The methodology developed opens up a new very promising way to develop extremely lifelike virtual mannequins that represent population groups that have common morphological characteristics.

INTRODUCTION

Using Digital Human Models in automotive sector is very common, especially for impact simulation and to determine reaches and spaces for the driver and the passengers. However, the models that are used have a standard morphology that is determined only with weight and height of the population under analysis. This results in geometric models with mean proportions and a distribution of generic body volumes that differ a lot from representing the morphological diversity of the population. This generates to simulated behaviors that are very different to the real ones. Therefore, the spaces designed according to these studies are not comfortable for  some users, thus jeopardizing the safety and the comfort of the people inside a vehicle.

Instituto de Biomecánica (IBV) has come up with the initial approach about the generation of three-dimensional digital human models in the driving posture and the study of the relationship between the morphological characteristics of the users in this posture and the anthropometric measurements obtained using traditional methods.

The research work conducted by the IBV was part of a bigger project sponsored and carried out by the JARI laboratory (Japanese Automotive Research Institute) of Japan, to investigate the mechanics of whiplash in women, by developing, validating and using finite element models (FEM) of the human body.

METHODOLOGY

The project was split up into three stages, as shown in the diagram of figure 1.

Figure 1. Diagram of the project

EXPERIMENTAL STAGE

The experimental stage involved the data collection, that took place in Madrid (Hospital Montepríncipe), according to the protocols developed by the IBV and agreed with the Japanese laboratory AIST (National Institute of Advanced Industrial Science and Technology).

Seven subjects took part in the study, men and women, specifically chosen for having different weights and heights, using the average European population as a reference.

1. Anthropometric measurements

The anthropometric characteristics of the participants is used to determine the sector of the population represented by each one of them.  Then the measurements of each participant were compared with a database of the Spanish female and male population.

Figure 2. Anthropometric measurements used for the comparative study
of the participants with the Spanish population.

2. 3D scanning of the whole body

To scan the body of the subjects in different postures, it was used the Vitus Smarl XXL scanner from Human Solutions was used, with a 1 mm level of accuracy and in only 12 seconds each one. The scanning volume is big enough for a person to fit in either standing up or seated in the driving posture.

The subjects were scanned in different standing postures and in various seated postures. For this an adjustable stool and two seats were used to simulate the driving posture. These seats were made specifically for the project, with the most common seat and back positions used in automobiles. The wood surface was covered with black fabric so that it didn’t appear in the scan and only the shape of the body was shown.

Figure 3. Left: Position of the seat in the scanning cabin to get the driving position.
Right: Normal position of the subject on the scanning platform.

Figure 4. Scanning postures included in the project protocol.

PROCESSING THE SCANS

The objective of this stage is to obtain a closed surface from the raw scan data, which contains noise and incomplete areas that are hidden to the cameras. To clean and correct these irregularities it has been used the Poisson reconstruction algorithms[1], “Surface fairing”[2] and the identification of the areas missing on the body, that are treated specifically in the harmonization of 3D scans (the adjustment step of the homologous model).

The result of the harmonization is a homologous model of each individual, which consists in a structured mesh and a skeleton, whose topological parameters are common among all the individuals, although their shape is unique and it is adjusted to the body of the scanned person.

This structure is characterized by a high local density with the right resolution to accurately produce the shapes of the body using the data from the raw 3D scans. Nevertheless, it is light enough to carry out a multivariate analysis of an extensive sample of the population, which facilitates the classification of individuals according to common morphological characteristics, thus obtaining digital human models that truly represent the “morphological” groups of the population.

To process the seated posture various scans of the same person in different postures are used so that incomplete information can be recovered. The method is based on a combination of two models, one is a deformation of the posture and the other is a variation of the 3D surface based in the body shape.

Figure 5. Left: Generation of the homodel in the seated posture. Right: Map of Hausdorff
distances between the original scan and the homodel.

[1] Kazhdan, M., Bolitho, M., & Hoppe, H. (2006, June). Poisson surface reconstruction. In Proceedings of the fourth Eurographics symposium on Geometry processing.

[2] Liepa, P. (2003). Filling holes in meshes. In Proceedings of the 2003 Eurographics/ACM SIGGRAPH symposium on Geometry processing (pp. 200-205). Eurographics Association.

DETERMINING THE MORPHOTYPE OF EACH PARTICIPANT

The objective of this stage is to analyze the representability of the participants in relation to the Spanish population. The study included:

♦ Determining the most representative percentile of each volunteer using the main anthropometric measurements: height, weight, length of the legs and the length of the arms.

♦ Variation of the body morphology compared with the average avatar representing the Spanish population. To this end, the work is carried out in an area that is defined by the modes of variation resulting from principal components analysis, where each principal component is a mode of variation of the human body.

Figure 6. Change in the first (height), second (mass) and third (length of the arms) modes
of variation of the Principal Components Analysis.

 CONCLUSIONS

The main contribution of the project to the use of 3D Scan technologies for the study of all-round safety in automotive sector has been the development of a methodology to obtain complete and simplified meshes of the human body in the driving posture. A procedure has also been developed to scan people in the driving posture using a scanner that is configured for the standing posture.

At the same time, the comparative study of the subjects with the Spanish population has been carried out, not only according to the traditional anthropometric measurements, but also an analysis of the main components of the three-dimensional shapes of the body has been used to characterize and compare the subjects with the population. These results help to identify and determine which population group represents each participant of the study, and therefore conclusions from other studies of impact simulation with those virtual models can be compared.

FUTURE WORK

The methodology developed in this project opens up a very promising way to produce extremely lifelike virtual mannequins that represent the population groups with common morphological characteristics.

Making more progress in the work with digital human models would involve making these models more complex, by incorporating the geometry of internal organs of the human body. Therefore, for example, the geometry of the spine or the thoracic cavity, obtained from a CAT scan or an MRI in the same driving posture might become part of more complex biomechanical models to study the effect of the impact on the skeleton or other internal organs.

ACKNOWLEDGMENTS

We would like to thank JARI and CHALMERS UNIVERSITY OF TECHNOLOGY for putting their trust in IBV and for kindly inviting us to take part in this stage of the project.

Read 6669 times Last modified on Tuesday, 19 September 2017 15:32



BUSCAR / FIND

SOCIAL MEDIA

facebook logo Twitter-X logo YouTube logo

NOSOTROS / ABOUT

 

INSTITUTO DE BIOMECÁNICA (IBV)

Universitat Politècnica de València •  Edificio 9C

Camino de Vera s/n  •  E-46022 Valencia  •  Spain

Tel. +34 96 111 11 70 •  +34 610 567 200

comunicacion@ibv.org  •  www.ibv.org

Copyright © 2024 Instituto de Biomecánica