Search Results

The purpose of this study was to develop a numerical analytical model of collinear low-speed bumper-to-bumper crashes and use the model to perform parametric studies of low-speed crashes and to estimate the severity of low-speed crashes that have already occurred. The model treats the car body as a rigid structure and the bumper as a deformable structure attached to the vehicle. The theory used in the model is based on Newton's Laws. The model uses an Impact Force-Deformation (IF-D) function to determine the impact force for a given amount of crush. The IF-D function used in the simulation of a crash that has already occurred can be theoretical or based on the measured force-deflection characteristics of the bumpers of the vehicles that were involved in the actual crash. The restitution of the bumpers is accounted for in a simulated crash through the rebound characteristics of the bumper system in the IF-D function.

A dynamic rollover test system (DRoTS) capable of simulating rollover crashes in a laboratory was designed for research use at the University of Virginia. The goal of the current study is to describe the system's capabilities and specifications as well as to explore the limitations of the system's ability to simulate rollover crashes. The test apparatus was designed to permit simulation of a single roof-to-ground interaction of a rollover crash with the potential to be modified for evaluation of pre-roof contact occupant motion. Special considerations were made to permit testing of both dummies and post-mortem human surrogates in both production vehicles and a parametric test buck. DRoTS permits vertical translation, pitch, and roll of the test vehicle while constraining longitudinal and lateral translations and yaw. The study details the ranges of test parameters capable with the DRoTS and evaluates the limitations of the system relative to rollover crash conditions.

Biomechanical data are often assumed to be doubly censored. In this paper, this assumption is evaluated critically for several previously published sets of data. Injury risk functions are compared using simple logistic regression and using survival analysis with 1) the assumption of doubly censored data and 2) the assumption of right-censored (uninjured specimens) and uncensored (injured) data. It is shown that the injury risk functions that result from these differing assumptions are not similar and that some experiments will require a preliminary assessment of data censoring prior to finalizing the experimental design. Some types of data are obviously doubly censored (e.g., chest deflection as a predictor of rib fracture risk), but many types are not left censored since injury is a force-limiting phenomenon (e.g., axial force as a predictor of tibia fracture). Guidelines for determining the censoring for various types of experiment are presented.

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.

Crash test dummies rely on biomechanical data from cadaver studies to biofidelically reproduce loading and predict injury. Unfortunately, it is difficult to obtain equivalent measurements of leg loading in a dummy and a cadaver, particularly for bending moments. A methodology is presented here to implant load cells in the tibia and fibula while minimally altering the functional anatomy of the two bones. The location and orientation of the load cells can be measured in all six degrees of freedom from post-test radiographs. Equations are given to transform tibial and fibular load cell measurements from a cadaver or dummy to a common leg coordinate frame so that test data can be meaningfully compared.

Previous lateral knee bending and shear tests have reported knee joint failure moments close to failure bending moments for the tibia and femur. Eight tibias, eight femurs and three knee joints were tested in lateral bending and two knee joints were tested in lateral shear. Seven previous studies on femur bending, five previous studies on tibia bending, two previous studies on knee joint bending, and one on shear were reviewed and compared with the current tests. All knee joint failures in the current study were either epiphysis fractures of the femur or soft tissue failures. The current study reports an average lateral failure bending moment for the knee joint (134 Nm SD 7) that is dramatically lower than that reported in the literature (284-351 Nm), that reported in the current study for the tibia (291 Nm SD 69) and for femur (382 Nm SD 103).

Rollover crashes are often difficult to reconstruct in detail because of their chaotic nature. Historically, vehicle speeds in rollover crashes have been calculated using a simple slide-to-stop formula with empirically derived drag factors. Roll rates are typically calculated in an average sense over the entire rollover or a segment of it in which vehicle roll angles are known at various positions. A unified model to describe the translational and rotational vehicle dynamics throughout the rollover sequence is lacking. We propose a pseudo-cylindrical model of a rolling vehicle in which the rotational and translational dynamics are coupled to each other based on the average frictional forces developed during ground contacts. We describe the model as pseudo-cylindrical because vertical motion is ignored but the ground reaction force is not constrained to act directly underneath the center of gravity of the vehicle.

Accident reconstructionists routinely rely on computer software to perform analyses. While there are a variety of software packages available to accident reconstructionists, many rely on custom spreadsheet-based applications for their analyses. Purchased packages provide an improved interface and the ability to produce sophisticated animations of vehicle motion but can be cost prohibitive. Pycrash is a free, open-source Python-based software package that, in its current state, can perform basic accident reconstruction calculations, automate data analyses, simulate single vehicle motion and, perform impulse-momentum based analyses of vehicle collisions. In this paper, the current capabilities of Pycrash are illustrated and its accuracy is assessed using matching PC-Crash simulations performed using PC-Crash.

Though rotation is thought to be the most common mechanism of foot and ankle injury in both automobile crashes and in everyday life, axial impact loading is considered responsible for most severe lower extremity injuries. In this study, dynamic axial impact tests were conducted on 92 isolated human lower limbs. The test apparatus delivered the impact via a pendulum-driven plate which intruded longitudinally to simulate the motion of the toepan in an automobile crash. Magneto-hydrodynamic (MHD) angular rate sensors fixed to the limbs measured ankle rotations during the impact event. Malleolar or fibula fractures, which are commonly considered to be caused by excessive ankle rotation, were present in 38% (12 out of 32) of the injured specimens. Ankle rotations in these tests were always within 10° of neutral at the time of peak axial load and seldom exceeded failure boundaries reported in the literature at any point during the impact event.

A follow-up case study on rollover testing with a single full-size sport utility vehicle (SUV) was conducted under controlled real-world conditions. The purpose of this study was to conduct a well-documented rollover event that could be utilized in evaluating various methods and techniques over the phases associated with rollover accidents. The phases documented and discussed, inherent to rollovers, are: pre-trip, trip, and rolling phases. With recent advances in technology, new devices and techniques have been designed which improve the ability to capture and document the unpredictable dynamic events surrounding vehicle rollovers. One such device is an inertial measurement unit (IMU), which utilizes GPS technology along with integrated sensors to report and record measured dynamic parameters real-time. The data obtained from a RT-4003 IMU device are presented and compared along with previous test data and methodology.

Calculating head accelerations and neck loading is essential for understanding and predicting head and neck injury. Most of the desired information cannot be directly measured in experiments with human volunteers. Achieving accurate results after applying the necessary transformations from remote measurements is difficult, particularly in the case of a head impact. The objective of this study was to develop a methodology for accurately calculating the accelerations at the center of gravity of the head and the loads and moments at the occipital condyles. To validate this methodology in a challenging test condition, twenty (20) human volunteers and a Hybrid III dummy were subjected to forehead impacts from a soccer ball traveling horizontally at speeds up to 11.5 m/s. The human subjects and the Hybrid III were instrumented with linear accelerometers and an angular rate sensor inside the mouth.

Photogrammetry, camera matching, and model-based image matching are commonly used techniques to analyze photographs and video for accident reconstruction and other forensic applications. Investigators are often tasked with taking measurements from photographs or determining speed from a video. All such calculations are based on fundamental geometric principles governing image projection inside a camera. Most treatments in the literature express the image projection equations in matrix notation rather than closed-form solutions. The purpose of this paper is to present a geometric derivation of the image projection equations in closed form that can be readily applied by a qualified investigator without the need for specialized software. In addition, a simple brute force optimization procedure is described to perform camera matching and model-based image matching. Examples are provided to demonstrate the method.

There are exposures of the body to accelerations in the lumbar and thoracic regions on a regular basis with everyday activities and exercises. The purpose of this study was to evaluate the response of the thoracic and lumbar regions in human volunteers subjected to vigorous activities. A total of 181 tests include twenty volunteers subjected to four test scenarios: “plopping” down in a seat, a vertical jump, a vertical drop while in a supine position, and a vertical drop while seated upright in a swing. Each of the latter three activities included three severity levels with drop heights ranging from 25 mm to 900 mm. Volunteers selected represent the anthropometry of the general population including males and females at a wide range of weights (54 to 99 kg), heights (150 to 191 cm), and ages (26 to 58 years old). Instrumentation for each volunteer included tri-axial accelerometers attached to custom-fit mounts that were secured around the lumbar and upper thoracic regions.

Vehicle dynamics in sideswipe collisions are markedly different from other types of collisions. Sideswipe collisions are characterized by prolonged sliding contact, often with very little structural deformation. An analytical model was developed to investigate the vehicle dynamics of sideswipe collisions. The vehicles were modeled as rigid bodies, and lateral interaction between the vehicles was modeled with a linear elastic spring. This linear spring was meant to represent the combined lateral stiffness of both vehicles before significant crush develops. Longitudinal interaction between the vehicles was modeled as frictional contact. In order to validate the model, seven (7) low speed (3 - 10 kph), shallow angle (15°) sideswipe collisions were staged with instrumented vehicles. These sideswipe collisions were characterized by long contact durations (∼ 1 s) and low accelerations (< 0.4 g's). The experimental collisions were also simulated with EDSMAC.

It has been proposed that low speed collisions in which the damage is isolated to the bumper systems can be reconstructed using data from customized quasistatic testing of the bumper systems of the involved vehicles. In this study, 10 quasistatic bumper tests were conducted on 7 vehicle pairs involved in front-to-rear collisions. The data from the quasistatic bumper tests were used to predict peak bumper force, vehicle accelerations, velocity changes, dynamic combined crush, restitution, and crash pulse time for a given impact velocity. These predictions were compared to the results measured by vehicle accelerometers in 12 dynamic crash tests at impact velocities of 2 - 10 mph. The average differences between the predictions using the quasistatic bumper data and the dynamic crash test accelerometer data were within 5% for bumper force, peak acceleration, and velocity change, indicating that the quasistatic bumper testing method had no systematic bias compared to dynamic crash testing.

A simple two-dimensional particle model was previously developed to calculate occupant ejection trajectories in rollover crashes. Model parameters were optimized using data from a dolly rollover test of a 1998 Ford Expedition in which five unbelted anthropomorphic test devices (ATDs) were completely ejected. In the present study, the model was further validated against a dolly rollover test of a 2004 Volvo XC90 in which three unbelted ATDs were completely ejected. The findings from the experimental testing were then compared to two real world rollover crashes with occupant ejections that were captured on video. The crashes were reconstructed by analyzing the video footage and aerial images of the crash sites. In both cases, the model was able to accurately match the observed trajectories of the ejected occupants, and the optimized model parameters were similar to the values obtained from the dolly rollover testing.

A simple two-dimensional particle model was developed to predict the airborne trajectory, landing point, tumbling distance, and rest position of an occupant ejected in a rollover crash. The ejected occupant was modeled as a projectile that was launched tangentially at a given radius from the center of gravity of the vehicle. The landing and tumbling phases of the ejection were modeled assuming a constant coefficient of friction between the occupant and the ground. Model parameters were optimized based on a dolly rollover test of a 1998 Ford Expedition in which five unbelted anthropomorphic test devices (ATDs) were completely ejected. A generalized vehicle dynamics model was also created assuming a constant translational deceleration and a prescribed roll rate function. Predictions using the generalized model were validated against the results of the full-scale rollover test to estimate the expected error when using the model in a real world situation.

This is the complete manuscript and replacement for SAE paper 2014-01-0482, which has been retracted due to incomplete content. This paper reports on 76 quasi-static tests conducted to investigate the behavior of road vehicle bumper systems. The tests are a quasi-static replication of real world low speed collisions. The tests represented front to rear impacts between various vehicles. Force and deflection were captured in order to quantify the stiffness characteristics of the bumper-to-bumper system. A specialized test apparatus was constructed to position and load bumper systems into each other. The purpose was to replicate or exceed damage that occurred in actual collisions. The fixture is capable of positioning the bumpers in various orientations and generates forces up to 50 kips. Various bumper-to-bumper alignments were tested including full overlap, lateral offset, and override/underride configurations.