722 Accident Reconstruction Simulation using Multi-body Child Human Model

Traffic accident reconstruction is a reverse dynamic problem, which requires hundreds of iterations to reconstruct the whole process of accident. However, in current pedestrian-vehicle accident reconstructions, it is difficult to quickly establish a pedestrian model based on specific cases, and it is hard to solve the contradiction between calculation accuracy and calculation time. In this paper, a personalized pedestrian customization method is proposed. First, the pedestrian structure is divided into independent modules according to obvious bony markers. For each independent module, multi-body (MB) model and finite element (FE) model are established, respectively. Then the appropriate modules are selected to form the whole hybrid pedestrian model. This method can customize the structure of pedestrian model according to the injury characteristics of pedestrians in specific accidents, and customize the parameters of pedestrian model according to the height and weight of pedestrians. The impact simulation tests are carried out on hybrid pedestrian models to verify the reliability of the models. The proposed method can effectively improve the modeling efficiency of pedestrian models and the reconstruction quality of pedestrian traffic accidents.

Download Full-text

HEAVY VEHICLE MULTI-BODY DYNAMIC SIMULATIONS TO ESTIMATE SKIDDING DISTANCE

The Baltic Journal of Road and Bridge Engineering ◽

10.3846/bjrbe.2018.384 ◽

2018 ◽

Vol 13 (1) ◽

pp. 23-33 ◽

Cited By ~ 2

Author(s):

Mahdieh ZAMZAMZADEH ◽

Ahmad Abdullah SAIFIZUL ◽

Rahizar RAMLI ◽

Ming Foong SOONG

Keyword(s):

Linear Regression ◽

Regression Model ◽

Coefficient Of Friction ◽

Heavy Vehicle ◽

Accident Reconstruction ◽

Heavy Vehicles ◽

The Road ◽

Gross Vehicle Weight ◽

Multi Body ◽

Body Dynamic

The skid mark is valuable for accident reconstruction as it provides information about the drivers’ braking behaviour and the speed of heavy vehicles. However, despite its importance, there is currently no mathematical model available to estimate skidding distance (SD) as a function of vehicle characteristics and road conditions. This paper attempts to develop a non-linear regression model that is capable of reliably predicting the skidding distance of heavy vehicles under various road conditions and vehicle characteristics. To develop the regression model, huge data sets were derived from complex heavy vehicle multi-body dynamic simulation. An emergency braking simulation was conducted to examine the skidding distance of a heavy vehicle model subject to various Gross Vehicle Weight (GVW) and vehicle speeds, as well as the coefficient of friction of the road under wet and dry conditions. The results suggested that the skidding distance is significantly affected by Gross Vehicle Weight, speeds, and coefficient of friction of the road. The improved non-linear regression model provides a better prediction of the skidding distance than that of the conventional approach thus suitable to be employed as an alternative model for skidding distance of heavy vehicles in accident reconstruction.

Download Full-text

50081 Development and Estimation of Multi-body Child Human Model(Biomechanics)

The Proceedings of the Asian Conference on Multibody Dynamics ◽

10.1299/jsmeacmd.2010.5._50081-1_ ◽

2010 ◽

Vol 2010.5 (0) ◽

pp. _50081-1_-_50081-8_

Author(s):

Takayuki Koizumi ◽

Nobutaka Tsujiuchi ◽

Azusa Nakai

Keyword(s):

Human Model ◽

Multi Body

Download Full-text

B-3 Development and Estimation of Multi-body Child Human Model

The Proceedings of Joint Symposium Symposium on Sports Engineering Symposium on Human Dynamics ◽

10.1299/jsmesports.2009.0_253 ◽

2009 ◽

Vol 2009 (0) ◽

pp. 253-258

Author(s):

Takayuki KOIZUMI ◽

Nobutaka TSUJIUCHI ◽

Azusa NAKAI ◽

Youichi MOTOMURA ◽

Yoshihumi NISHIDA

Keyword(s):

Human Model ◽

Multi Body

Download Full-text

A multi-body human model for rear-impact simulation

Proceedings of the Institution of Mechanical Engineers Part D Journal of Automobile Engineering ◽

10.1243/09544070jauto985 ◽

2009 ◽

Vol 223 (5) ◽

pp. 623-638 ◽

Cited By ~ 6

Author(s):

S Himmetoglu ◽

M Acar ◽

K Bouazza-Marouf ◽

A Taylor

Keyword(s):

Human Body ◽

Seat Belt ◽

Impact Response ◽

Human Model ◽

Human Body Model ◽

Head Restraint ◽

Impact Simulation ◽

Body Model ◽

Rear Impact ◽

Multi Body

This paper presents the validation of a 50th-percentile male multi-body human model specifically developed for rear-impact simulation. The aim is to develop a biofidelic model with the simplest architecture that can simulate the interaction of the human body with the seat during rear impact. The model was validated using the head-and-neck and torso responses of seven volunteers from the Japanese Automobile Research Institute sled tests, which were performed at an impact speed of 8km/h with a rigid seat and without head restraint and seat belt. The results indicate that the human-body model can effectively mimic the rear-impact response of a 50th-percentile male with a good level of accuracy and has the potential to predict whiplash injury.

Download Full-text