Search Results

Current state-of-the-art vehicles implement pedestrian protection features that rely on pedestrian detection sensors and algorithms to trigger when impacting a pedestrian. During the development phase, the vehicle must “learn” to discriminate pedestrians from the rest of potential impacting objects. Part of the training data used in this process is often obtained in physical tests utilizing legform impactors whose external biofidelity is still to be evaluated. This study uses THUMS as a reference to assess the external biofidelity of the most commonly used impactors (Flex-PLI, PDI-1 and PDI-2). This biofidelity assessment was performed by finite element simulation measuring the bumper beam forces exerted by each surrogate on a sedan and a SUV. The bumper beam was divided in 50 mm sections to capture the force distribution in both vehicles. This study, unlike most of the pedestrian-related literature, examines different impact locations and velocities.

The goal of this study was to develop injury probability functions for the leg bending moment and MCL (Medial Collateral Ligament) elongation of the Flexible Pedestrian Legform Impactor (Flex-PLI) based on human response data available from the literature. Data for the leg bending moment at fracture in dynamic 3-point bending were geometrically scaled to an average male using the standard lengths obtained from the anthropometric study, based on which the dimensions of the Flex-PLI were determined. Both male and female data were included since there was no statistically significant difference in bone material property. Since the data included both right censored and uncensored data, the Weibull Survival Model was used to develop a human leg fracture probability function.

This paper presents a technique for developing corridors from individual subject responses contained in experimental biomechanical data sets. Force-deflection response is used as an illustrative example. The technique begins with a method for averaging human subject force-deflection responses in which curve shape characteristics are maintained and discontinuities are avoided. Individual responses sharing a common characteristic shape are averaged based upon normalized deflection values. The normalized average response is then scaled to represent the given data set using the mean peak deflection value associated with the set of experimental data. Finally, a procedure for developing a corridor around the scaled normalized average response is presented using standard deviation calculations for both force and deflection.

This paper presents an analysis of the displacement measurement of the Hybrid III 50th percentile male dummy chest in quasistatic and dynamic loading environments. In this dummy, the sternal chest deformation is typically characterized using a sliding chest potentiometer, originally designed to measure inward deflection in the central axis of the dummy chest. Loading environments that include other modes of deformation, such as lateral translations or rotations, can create a displacement vector that is not aligned with this sensitive axis. To demonstrate this, the dummy chest was loaded quasistatically and dynamically in a series of tests. A string potentiometer array, with the capability to monitor additional deflection modes, was used to supplement the measurement of the chest slider.

Studies show that the pedestrian population at high risk of injury consists of both young children and adults. The goal of this study is to gain understanding in the mechanisms that lead to injuries for children and adults. Multi-body pedestrian human models of two specific anthropometries, a 6year-old child and a 50th percentile adult male, are applied. A vehicle model is developed that consists of a detailed rigid finite element mesh, validated stiffness regions, stiff structures underlying the hood and a suspension model. Simulations are performed in a test matrix where anthropometry, impact speed and impact location are variables. Bumper impact occurs with the tibia of the 50th percentile adult male and with the thigh of the 6-year-old child. The head of a 50th percentile male impacts the lower windshield, while the 6-year-old child's head impacts the front part of the hood.

This research investigates the variation of pedestrian stance in pedestrian-automobile impact using a validated multi-body vehicle and human model. Detailed vehicle models of a small family car and a sport utility vehicle (SUV) are developed and validated for impact with a 50th percentile human male anthropometric ellipsoid model, and different pedestrian stances (struck limb forward, feet together, and struck limb backward) are investigated. The models calculate the physical trajectory of the multi-body models including head and torso accelerations, as well as pelvic force loads. This study shows that lower limb orientation during a pedestrian-automobile impact plays a dominant role in upper body kinematics of the pedestrian. Specifically, stance has a substantial effect on the subsequent impacts of the head and thorax with the vehicle. The variation in stance can change the severity of an injury incurred during an impact by changing the impact region.

This paper presents a computational study of the effects of three parameters on the resulting thoracic injury criteria in side impacts. The parameters evaluated are a) door velocity-time (V-t) profile, b) door interior padding modulus, and c) initial door-to-occupant offset. Regardless of pad modulus, initial offset, or the criterion used to assess injury, higher peak door velocity is shown to correspond with more severe injury. Injury outcome is not, however, found to be sensitive to the door velocity at the time of first occupant contact. A larger initial offset generally is found to result in lower injury, even when the larger offset results in a higher door velocity at occupant contact, because the increased offset results in contact later in the door V-t profile - closer to the point at which the door velocity begins to decrease. Cases of contradictory injury criteria trends are identified, particularly in response to changes in the pad modulus.

Some rollover testing methodologies require specification of vehicle kinematic parameters including travel speed, vertical velocity, roll rate, and pitch angle, etc. at the initiation of vehicle to ground contact, which have been referred to as touchdown conditions. The complexity of the vehicle, as well as environmental and driving input characteristics make prediction of realistic touchdown conditions for rollover crashes, and moreover, identification of parameter sensitivities of these characteristics, is difficult and expensive without simulation tools. The goal of this study was to study the sensitivity of driver input on touchdown parameters and the risk of rollover in cases of steering-induced soil-tripped rollovers, which are the most prevalent type of rollover crashes. Knowing the range and variation of touchdown parameters and their sensitivities would help in picking realistic parameters for simulating controlled rollover tests.

Fatalities resulting from vehicle rollover events account for over one-third of all U.S. motor vehicle occupant fatalities. While a great deal of research has been directed towards the rollover problem, few studies have attempted to determine the sensitivity of occupant injury risk to variations in the vehicle (roof strength), crash (kinematic conditions at roof-to-ground contact), and occupant (anthropometry, position and posture) parameters that define the conditions of the crash. A two-part computational study was developed to examine the sensitivity of injury risk to changes in these parameters. The first part of this study, the Crash Parameter Sensitivity Study (CPSS), demonstrated the influence of parameters describing the vehicle and the crash on vehicle response using LS-DYNA finite element (FE) simulations.

The goal of this study was to develop and validate a finite element (FE) model of the Polar-II pedestrian dummy. An upper body model consisting of the head, neck, shoulder, thorax, and abdomen was coupled with a previously validated model of the lower limb The viscoelastic material properties of the dummy components were determined from dynamic compression tests of shoulder urethane, shoulder rubber and abdominal foam. For validation of the entire upper body, the model was compared with NHTSA response requirements for their advanced frontal dummy (Thor) including head and neck pendulum tests as well as ribcage and abdominal impact tests. In addition, the Polar-II full body FE model was subjected to simulated vehicle-pedestrian impacts that recreated published experiments. Simulated head and pelvis accelerations as well as upper body trajectories reasonably reproduced the experiment.

A finite element (FE) model that can predict impact response and injuries to a human pelvis and lower limb was developed in PAM-CRASH™ by accurately representing human anatomical structures. In our previous study, three-dimensional (3D) geometry of the thigh, leg and knee joint was developed based on MRI scans from a human volunteer. 3D geometry of a bony pelvis created in this study was based on CT scans from a Post Mortem Human Subject (PMHS). The model was validated using published quasi-static and dynamic test results with human pelves and lower limbs. The thigh and leg models were validated against recently published dynamic 3-point bending test results with off-center loading. The validation results showed that this model can reproduce force-deflection and moment-deflection responses of a human thigh and leg in various loading conditions along with average force and moment at fracture.

Pedestrian impact protection has been a growing area of research over the past twenty or more years. The results from many studies have shown the importance of providing protection to vulnerable road users as a means of reducing roadway fatalities. Most of this research has focused on the vehicle fleet as a whole in datasets that are dominated by passenger cars (cars). Historically, the influence of vehicle body type on injury distribution patterns for pedestrians has not been a primary research focus. In this study we used the Pedestrian Crash Data Study (PCDS) database of detailed pedestrian crash investigations to identify how injury patterns differ for pedestrians struck by light trucks, vans, and sport utility vehicles (LTVs) from those struck by cars. AIS 2+ and 3+ injuries for each segment of vehicles were mapped back to both the body region of the pedestrian injured and the vehicle source linked to that injury in the PCDS database.

This research uses simulations of vehicle-pedestrian collisions to determine if the risk of pedestrian head injury is greater from impact with the vehicle or from impact with the ground, and to determine the influence of vehicle speed, vehicle type, and pedestrian stance on the injury risk. Five speeds, two vehicle types and four pedestrian stances are examined. In addition, a smaller set of simulations is included to determine the influence of body orientation just prior to ground impact. As anticipated, risk of head injury from both the vehicle and the ground tends to increase with vehicle speed, but injury risk from the ground is less predictable. At lower speeds, the vehicle tends to pose a greater risk of injury than does the ground, while at higher speeds the probability of injury from both the vehicle and ground is typically very large.

The evaluation of pedestrian safety performance of vehicles required by regulations and new car assessment programs (NCAPs) have been conducted. However, the behavior of a pedestrian in an actual car-pedestrian accident is complex. In order to investigate injuries to the pedestrian lower body, the biofidelity of the lower limb and the pelvis of a pedestrian dummy called the POLAR II had been improved in past studies to develop a prototype of the next generation dummy called the POLAR III. The biofidelity of the thigh and the leg of the POLAR III prototype has been evaluated by means of 3-point bending. However, the inertial properties of these parts still needed to be adjusted to match those of a human. The biofidelity of the pelvis of the POLAR III prototype has been evaluated in lateral compression. Although the experiment using PMHSs (Post Mortem Human Subjects) was conducted in dynamic condition, the dummy tests were performed only in quasi-static condition.

Multibody human models are widely used to investigate responses of human during an automotive crash. This study aimed to validate a commercially available multibody human body model against response corridors from volunteer tests conducted by Naval BioDynamics Laboratory (NBDL). The neck model consisted of seven vertebral bodies, and two adjacent bodies were connected by three orthogonal linear springs and dampers and three orthogonal rotational springs and dampers. The stiffness and damping characteristics were scaled up or down to improve the biofidelity of the neck model against NBDL volunteer test data because those characteristics were encrypted due to confidentiality. First, sensitivity analysis was performed to find influential scaling factors among the entire set using a design of experiment.

The high frequency of fatal head injuries is one of the important issues in traffic safety, and Traumatic Brain Injuries (TBIs) without skull fracture account for approximately half of them in both occupant and pedestrian crashes. In order to evaluate vehicle safety performance for TBIs in these crashes using anthropomorphic test dummies (ATDs), a comprehensive injury criterion calculated from the rotational rigid motion of the head is required. While many studies have been conducted to investigate such an injury criterion with a focus on diffuse brain injuries in occupant crashes, there have been only a limited number of studies focusing on pedestrian impacts. The objective of this study is to develop a comprehensive injury criterion based on the rotational rigid body motion of the head suitable for both occupant and pedestrian crashes.

Japanese accident statistics show that despite the decreasing trend of the overall traffic fatalities, more than 1,000 pedestrians are still killed annually in Japan. One way to develop further understanding of real-world pedestrian accidents is to reconstruct a variety of accident scenarios dynamically using computational models. Some of the past studies done by the authors' group have used a simplified vehicle model to investigate pedestrian lower limb injuries. However, loadings to the upper body also need to be reproduced to predict damage to the full body of a pedestrian. As a step toward this goal, this study aimed to develop a simplified vehicle model capable of reproducing pedestrian full-body kinematics and pelvis and lower limb injury measures. The simplified vehicle model was comprised of four parts: windshield, hood, bumper and lower part of the bumper. Several different models were developed using different combinations of geometric and stiffness representation.

Some rollover test methods, which impose a touchdown condition on a test vehicle, have been developed to study vehicle crashworthiness and occupant protection in rollover crashes. In ground-tripped rollover crashes, speed, steering maneuver, braking, vehicle inertial and geometric properties, topographical and road design characteristics, and soil type can all affect vehicle touchdown conditions. It is presumed that while there may be numerous possible combinations of kinematic metrics (velocity components and orientation) at touchdown, there are also numerous combinations of metrics that are not likely to occur in rollover crashes. To determine a realistic set of touchdown conditions to be used in a vehicle rollover crash test, a lateral deceleration sled-based non-destructive rollover initiation test system (RITS) with a fully programmable deceleration pulse is in development.

The goal of this study was to evaluate the biofidelity of the three computational surrogates (GHBMC model, WorldSID model, and the FTSS ES-2re model) under the side impact rigid wall sled test condition. The responses of the three computational surrogates were compared to those of post mortem human surrogate (PMHS) and objectively evaluated using the correlation and analysis (CORA) rating method. Among the three computational surrogates, the GHBMC model showed the best biofidelity based on the CORA rating score (GHBMC =0.65, WorldSID =0.57, FTSS ES-2re =0.58). In general, the response of the pelvis of all the models showed a good correlation with the PMHS response, while the response of the shoulder and the lower extremity did not. In terms of fracture prediction, the GHBMC model overestimated bone fracture.

The validity of evaluating FlexPLI peak injury measures has been shown by the correlation of the peak measures between a human FE model and a FlexPLI FE model. However, comparisons of tibia bending moment time histories (BMTHs) between these models show that the FlexPLI model exhibits a higher degree of oscillatory behavior than the human model. The goal of this study was to identify potential improvements to the FlexPLI such that the legform provides more biofidelic tibia BMTHs at the normal standing height. Impact simulations using a human FE model and a FlexPLI FE model were conducted against simplified vehicle models to compare tibia BMTHs. The same series of impact simulations were conducted using the FlexPLI models that incorporated potential measures to identify measures effective for further enhancement of the biofidelity. An additional analysis was also conducted to investigate the key factor for minimizing the oscillation of the tibia BMTH.