Viewing 1 to 29 of 29
Technical Paper
Matthew E. O'Kelly, Houssam Abbas, Sicun Gao, Shinpei Kato, Shinichi Shiraishi, Rahul Mangharam
Abstract Autonomous vehicles (AVs) have already driven millions of miles on public roads, but even the simplest scenarios have not been certified for safety. Current methodologies for the verification of AV’s decision and control systems attempt to divorce the lower level, short-term trajectory planning and trajectory tracking functions from the behavioral rules-based framework that governs mid-term actions. Such analysis is typically predicated on the discretization of the state space and has several limitations. First, it requires that a conservative buffer be added around obstacles such that many feasible plans are classified as unsafe. Second, the discretized controllers modeled in this analysis require several refinement steps before being implementable on an actual AV, and typically do not allow the specification of comfort-related properties on the trajectories. Consumer-ready AVs use motion planning algorithms that generate smooth trajectories.
Journal Article
Devin SJ Caplow-Munro, Helen Loeb, Venk Kandadai, Flaura Winston
Abstract Inadequate situation awareness and response are increasingly recognized as prevalent critical errors that lead to young driver crashes. To identify and assess key indicators of young driver performance (including situation awareness), we previously developed and validated a Simulated Driving Assessment (SDA) in which drivers are safely and reproducibly exposed to a set of common and potentially serious crash scenarios. Many of the standardized safety measures can be calculated in near real-time from simulator variables. Assessment of situation awareness, however, largely relies on time-consuming data reduction and video coding. Therefore, the objective of this research was to develop a near real-time automated method for analyzing general direction and location of driver's gaze in order to assess situation awareness.
Technical Paper
Flaura Winston, Catherine McDonald, Venk Kandadai, Zachary Winston, Thomas Seacrist
Abstract Driving simulators offer a safe alternative to on-road driving for the evaluation of performance. In addition, simulated drives allow for controlled manipulations of traffic situations producing a more consistent and objective assessment experience and outcome measure of crash risk. Yet, few simulator protocols have been validated for their ability to assess driving performance under conditions that result in actual collisions. This paper presents results from a new Simulated Driving Assessment (SDA), a 35- to-40-minute simulated assessment delivered on a Real-Time® simulator. The SDA was developed to represent typical scenarios in which teens crash, based on analyses from the National Motor Vehicle Crash Causation Survey (NMVCCS). A new metric, failure to brake, was calculated for the 7 potential rear-end scenarios included in the SDA and examined according two constructs: experience and skill.
Technical Paper
Helen S. Loeb, Thomas Seacrist, Catherine McDonald, Flaura Winston
Abstract Driving simulators provide a safe, highly reproducible environment in which to assess driver behavior. Nevertheless, data reduction to standardized metrics can be time-consuming and cumbersome. Further, the validity of the results is challenged by inconsistent definitions of metrics, precluding comparison across studies and integration of data. No established tool has yet been made available and kept current for the systematic reduction of literature-derived safety metrics. The long term goal of this work is to develop DriveLab, a set of widely applicable routines for reducing simulator data to expert-approved metrics. Since Matlab™ is so widely used in the research community, it was chosen as a suitable environment. This paper aims to serve as a case study of data reduction techniques and programming choices that were made for simulator analysis of a specific research project, the Simulated Driving Assessment.
Technical Paper
Sijia Zhang, Kristen J. Nicholson, Jenell R. Smith, Taylor M. Gilliland, Peter P. Syré, Beth A. Winkelstein
Both traumatic and slow-onset disc herniation can directly compress and/or chemically irritate cervical nerve roots, and both types of root injury elicit pain in animal models of radiculopathy. This study investigated the relative contributions of mechanical compression and chemical irritation of the nerve root to spinal regulation of neuronal activity using several outcomes. Modifications of two proteins known to regulate neurotransmission in the spinal cord, the neuropeptide calcitonin gene-related peptide (CGRP) and glutamate transporter 1 (GLT-1), were assessed in a rat model after painful cervical nerve root injuries using a mechanical compression, chemical irritation or their combination of injury. Only injuries with compression induced sustained behavioral hypersensitivity (p≤0.05) for two weeks and significant decreases (p<0.037) in CGRP and GLT-1 immunoreactivity to nearly half that of sham levels in the superficial dorsal horn.
Technical Paper
Kristen J. Nicholson, Julia C. Quindlen, Beth A. Winkelstein
Cervical nerve roots are susceptible to compression injuries of various durations. The duration of an applied compression has been shown to contribute to both the onset of persistent pain and also the degree of spinal cellular and molecular responses related to nociception. This study investigated the relationship between peripherally evoked activity in spinal cord neurons during a root compression and the resulting development of axonal damage. Electrically evoked spikes were measured in the spinal cord as a function of time during and after (post-compression) a 15 minute compression of the C7 nerve root. Compression to the root significantly (p=0.035) reduced the number of spikes that were evoked over time relative to sham. The critical time for compression to maximally reduce evoked spikes was 6.6±3.0 minutes. A second study measured the post-compression evoked neuronal activity following compression applied for a shorter, sub-threshold time (three minutes).
Technical Paper
Raymond D. Hubbard, Kyle P. Quinn, Joan J. Martinez, Beth A. Winkelstein
Rapid neck motions can load cervical nerve roots and produce persistent pain. This study investigated the cellular basis of radicular pain and mechanical implications of tissue loading rate. A range of peak loads was applied in an in vivo rat model of dorsal root compression, and mechanical allodynia (i.e., pain) was measured. Axonal damage and nociceptive mediators were assessed in the axons and cell bodies of compressed dorsal roots in separate groups of rats at days 1 and 7 after injury. In the day 7 group, damage in the compressed axons, evaluated by decreased heavy chain neurofilament immunoreactivity, was increased for compressions above a load of 34.08 mN, which is similar to the load-threshold for producing persistent pain in that model.
Technical Paper
Hongwei Hsiao, Norman I. Badler, Don B. Chaffin, King H. Yang, John F. Lockett
Digital human modeling (DHM) progress worldwide will be much faster and cohesive if the diverse community now developing simulations has a global blueprint for DHM, and is able to work together efficiently. DHM developers and users can save time by building on each other's work. This paper highlights a panel discussion on DHM goals and strategic plans for the next decade to begin formulating the international blueprint. Four subjects are chosen as the starting points: (1) moving DHM into the public safety and internet arenas, (2) role of DHM in computer assisted surgery and automotive safety, (3) DHM in defense applications, and (4) DHM to improve workplace ergonomics.
Technical Paper
Kyle P. Quinn, Kathryn E. Lee, Chukwunyere C. Ahaghotu, Beth A. Winkelstein
While studies have demonstrated the cervical facet capsule is at risk for tensile injury during whiplash, the relationship between joint loading, changes in the capsule's structure, and pain is not yet fully characterized. Complementary approaches were employed to investigate the capsule's structure-function relationship in the context of painful joint loading. Isolated C6/C7 facet joints (n=8) underwent tensile mechanical loading, and measures of structural modification were compared for two distraction magnitudes: 300 µm (PV) and 700 µm (SV). In a matched in vivo study, C6/C7 facet joints (n=4) were harvested after the same SV distraction and the tissue was sectioned to analyze collagen fiber organization using polarized light microscopy. Laxity following SV distraction (7.30±3.01%) was significantly greater (p<0.001) than that produced following PV distraction (0.99±0.44%).
Technical Paper
Jan M. Allbeck, Norman I. Badler
Computational ergonomic analyses are often laboriously tested one task at a time. As digital human models improve, we can partially automate the entire analysis process of checking human factors requirements or regulations against a given design. We are extending our Parameterized Action Representation (PAR) to store requirements and its execution system to drive human models through required tasks. Databases of actions, objects, regulations, and digital humans are instantiated into PARs and executed by analyzers that simulate the actions on digital humans and monitor the actions to report successes and failures. These extensions will allow quantitative but localized design assessment relative to specific human factors requirements
Technical Paper
Norman I. Badler, Jan Allbeck, Seung-Joo Lee, Richard J. Rabbitz, Timothy T. Broderick, Kevin M. Mulkern
The earliest Digital Human Modeling systems were non-interactive analysis packages with crude graphics. Next generation systems added interactivity and articulated kinematic human models. The newest systems use real-time computer graphics, deformable figures, motion controllers, and user interfaces. Our long-term goal is to free the user as much as possible from interactive human model manipulation through direct understanding and execution of task instructions. We present a next generation DHM testbed that includes a scriptable interface, real-time collision-avoidance reach, empirical joint motion models, a versatile locomotion engine, motion capture and synthetic motion blends and combinations, and a smooth skinned scalable human model.
Technical Paper
Liming Zhao, Ying Liu, Norman I. Badler
Simulating human reach is still challenging when considering complex interactions with the environment. Standard approaches involve inverse kinematics (IK) methods and usually require a complete but exponential cost search in configuration space. In ergonomic applications, both “naturalness” and interactive performance are important. We describe a real-time, collision-free, sternum-rooted IK solution for an articulated human figure based on motion capture data, human strength models, and multi-joint coordination functions. Movement paths are discovered through spatial search in a partitioned workspace. The system generates natural collision-free reach motions in real-time. The resulting animations and statistics demonstrate the efficacy of this approach.
Technical Paper
Kathryn E. Lee, Martin B. Davis, Roanne M. Mejilla, Beth A. Winkelstein
While extensive research points to mechanical injury of the cervical facet joint as a mechanism of whiplash injury, findings remain speculative regarding its potential for causing pain. The purpose of this study was to examine the relationship between facet joint distraction, capsular ligament strain, cellular nociceptive responses, and pain. A novel rat model of in vivo facet joint injury was used to impose C6/C7 joint distraction in separate studies of subcatastrophic and physiologic vertebral distraction, as well as sham procedures. A common clinical measure of behavioral hypersensitivity (allodynia) was measured for 14 days after injury, as quantification of resulting pain symptoms. Also, on day 14, spinal activation of microglia and astrocytes was quantified to examine the potential role of glial activation as a physiologic mechanism of facet-mediated painful injury. Vertebral distractions of 0.90±0.53 mm across the rat facet joint reliably produced symptoms of persistent pain.
Technical Paper
Kristy B. Arbogast, Shresta Mari-Gowda, Michael J. Kallan, Dennis R. Durbin, Flaura K. Winston
Little is known about the mechanism of pelvic injury in the pediatric population, an age range over which the pelvis undergoes tremendous structural change. We hypothesize that these structural changes influence pelvic fracture injury mechanisms. A probability sample of children under age 16 years in crashes were enrolled in an on-going crash surveillance system which links insurance claims data to telephone survey and crash investigation data. 15,725 children in side impact collisions were studied. Risk of pelvic fracture in side impact collisions was estimated and factors associated with these injuries were identified. Eight cases were examined using in-depth investigation to identify the injury mechanisms. Of our study sample, 0.10% of children suffered a pelvic fracture. The typical child with a pelvic fracture was a 12-15 year old female front row occupant of a passenger car involved in a struck side collision with intrusion.
Technical Paper
Gary M. Crosbie, Erica Perry Murray, David R. Bauer, Hyuk Kim, Seungdoo Park, John M. Vohs, Raymond J. Gorte
To be practical, auxiliary power units (APUs) should operate on the same fuels that the internal combustion engine (ICE) uses for vehicle propulsion. Solid oxide fuel cells (SOFCs) have previously been shown to be able to convert the chemical energy of certain room-temperature-liquid hydrocarbon fuels (toluene and synthetic diesel fuel) to electricity by direct oxidation. Because such SOFCs operate without reformers, the systems based on these SOFCs are expected to be compact. To work with existing infrastructure fuels, the cells must be able to tolerate typical contaminants such as sulfur that are found in the everyday fuels. In this paper, we report on recent laboratory results that show direct oxidation SOFCs with ceria-copper anodes can provide at least 2 hours operation in the presence of 200 ppm sulfur in the fuel. Also, a laboratory cell has been run for 12 hours on regular unleaded gasoline.
Technical Paper
Gary M. Crosbie, Erica Perry Murray, David R. Bauer, Hyuk Kim, Seungdoo Park, John M. Vohs, Raymond J. Gorte
To meet the increasing electrical power demands for advanced internal combustion engine (ICE) vehicles, auxiliary power units (APUs) are of growing interest. Fuel cell based APUs offer the potential for high chemical-to-electrical conversion efficiency with low noise and low emissions. It has recently been shown that solid oxide fuel cells (SOFCs) can be used to directly convert the chemical energy of liquid hydrocarbon fuels to electricity. Because the combustion reaction takes place by direct oxidation of vaporized fuel at the fuel cell anode, the expectation exists for development of compact, reformerless APUs that can operate on the same fuel that the ICE uses for vehicle propulsion. Critical issues for the transportation SOFC-APU applications are fast start-up and the need to survive extensive thermal cycling.
Technical Paper
Reid T. Miller, Douglas H. Smith, Xiaohan Chen, Bai-Nan Xu, Matt Leoni, Masahiro Nonaka, David F. Meaney
Validating a traumatic brain injury finite element model is often limited by a lack of extensive animal injury data that may be used to examine the conditions under which the model is accurate. Given that most published reports specify only general descriptions of injury, this study examined potential evaluation strategies and assessed the ability of a finite element model to simulate the general descriptions of injury in an animal model. The results of this study showed that 1) the results from a simplified finite element model could estimate trends that were similar to the injury patterns observed in a set of animal experiments, 2) a parameter (Z parameter), which quantified the comparison process between computational and animal data, estimated trends that would help in the model evaluation process, and 3) a more complete evaluation process would occur if multiple testing methods were included in the evaluation procedure.
Technical Paper
Michael T. Prange, Gyorgy Kiralyfalvi, Susan S. Margulies
Head injury is the most common cause of death and acquired disability in childhood. We seek to determine the influence of brain mechanical properties on inertial pediatric brain injury. Large deformation material properties of porcine pediatric and adult brain tissue were measured and represented by a first-order Ogden hyperelastic viscoelastic constitutive model. A 3-D finite element mesh was created of a mid-coronal slice of the brain and skull of a human adult and child (2 weeks old). Three finite element models were constructed: (1) a pediatric mesh with pediatric brain properties, (2) a pediatric mesh with adult tissue properties, and (3) an adult mesh with adult tissue properties. The skull was modeled as a rigid solid and an angular acceleration was applied in the coronal plane with center at C4/C5. The brain is assumed to be homogeneous and isotropic.
Technical Paper
Allison C. Bain, David F. Meaney
In vivo, tissue-level, mechanical thresholds for axonal injury in the guinea pig optic nerve were determined by comparing morphological injury to estimated in vivo tissue strain. The right optic nerve of adult male guinea pigs was stretched to one of seven ocular displacement levels. Morphological injury was detected three days post-stretch with neurofilament immunohistochemical staining (NF68). A companion set of in situ experiments was used to determine the empirical relationship between ocular displacement and optic nerve stretch. Logistics regression analysis, combined with sensitivity and specificity measures and receiver operating characteristic (ROC) curves were then used to predict strain thresholds for axonal injury. From this analysis, we determined three Lagrangian strain- based thresholds for morphological damage to the guinea pig white matter.
Technical Paper
Richard Grace, Alberto Guzman, James Staszewski, David F. Dinges, Melissa Mallis, Bethany A. Peters
Carnegie Mellon Driving Research Center, together with ISIM, is presently involved in the design and development of an Advanced Human Factors Research and Driving Training Research Facility. The facility has been designed to address human factors issues and driver training issues. Human factors interests include developing countermeasures for fatigue and driver/vehicle interface issues. Driver training issues include validating the usefulness of simulators for driver training, developing effective curricula and investigating simulator fidelity needed for effective training. A key component of the facility is the Carnegie Mellon TruckSim that will be capable of simulating a variety of commercial and emergency vehicles using interchangeable cabs mounted to a common motion platform. TruckSim's modular configuration will allow for rapid and cost effective design of experiments and training scenarios. A first research program to evaluate fatigue countermeasures is presented as an example.
Technical Paper
Reid T. Miller, Susan S. Margulies, Matt Leoni, Masahiro Nonaka, Xiaohan Chen, Douglas H. Smith, David F. Meaney
Traumatic brain injury finite element analyses have evolved from crude geometric representations of the skull and brain system into sophisticated models which take into account distinct anatomical features. However, two distinct finite element modeling approaches have evolved to account for the relative motion that occurs between the skull and cerebral cortex during traumatic brain injury. The first and most common approach assumes that the relative motion can be estimated by representing the cerebrospinal fluid inside the subarachnoid space as a low shear modulus, virtually incompressible solid. The second approach assumes that the relative motion can be approximated by defining a frictional interface between the cerebral cortex and dura mater. This study presents data from an experimental model of traumatic brain injury coupled with finite element analyses to evaluate the modeling approach's ability to predict specific forms of traumatic brain injury.
Technical Paper
David I. Shreiber, Allison C. Bain, David F. Meaney
A finite element model of cerebral contusion in the rat was developed and compared to experimental injury maps demonstrating blood-brain barrier (BBB) breakdown. The model was exercised at the nine unique loading conditions used experimentally. Logistic regressions of four variables, maximum principal logarithmic strain (LEP), maximum principal stress (SP), strain energy density (SEN), and von Mises stress (MIS) demonstrated highly significant confidence in the prediction of the 50th percentile values (chi-squared, p<0.00001). However, only values for LEP were invariant across loading conditions. These results suggest that the BBB is most sensitive to LEP, and that breakdown occurs above a strain of 0.188 +/- 0.0324.
Technical Paper
Kristy B. Arbogast, Susan S. Margulies
Computational modeling is a potentially powerful tool to provide information about the mechanisms of traumatic brain injury. In order to ensure that the estimates calculated by these computer models provide the most useful information, it is essential that these models contain accurate central nervous system (CNS) tissue properties. Previous material property measurements lack strict control over crucial experimental parameters that may influence material properties and tail to examine any regional variation in the measured response. To address these issues, we measured the material response of two regions of the CNS, the brainstem and the cerebrum. Specifically, adult porcine tissue was subjected to high loading rate mechanical deformation using a custom designed oscillatory shear device. Complex shear moduli were calculated over a range of frequencies (20-200 Hz) at two engineering strain amplitudes (2.5%, and 5.0%).
Technical Paper
T. Egami, W. Dmowski, R. Brezny
The use of cerium oxide as an oxygen storage component in automotive three-way catalysis has been well established. More recently the need to thermally stabilize these materials against deactivation at higher temperatures has focused attention on doping of the ceria with a wide range of metal oxides. The role of these dopants in the stabilization mechanisms for ceria is not completely understood as they must perform the complex role of sintering inhibitor while promoting oxygen storage and release. The scattering of pulsed neutrons produced by a spallation source coupled with the Fourier analysis provides a powerful method to characterize the local atomic structure of complex systems such as mixed oxides. We demonstrate that by using this method it is possible to obtain valuable information on the local atomic structure of the CeO2/ZrO2 catalyst support that cannot be attained by the conventional diffraction methods.
Technical Paper
Kristy Bittenbender Arbogast, David F. Meaney, Lawrence E. Thibault
Experimental tests using porcine brainstem samples were performed on a custom designed stress relaxation shear device. Tests were performed dynamically at strain rates >1 s−1, to three levels of peak strain (2.5%-7.5%). The directional dependence of the material properties was investigated by shearing both parallel and transverse to the predominant direction of the axonal fibers. Quasi-linear viscoelastic theory was used to describe the reduced relaxation response and the instantaneous elastic function. The time constants of the reduced relaxation function demonstrate no directional dependence; however, the relative magnitude of the exponential functions and the parameter representing the final limiting value are significantly different for each direction. The elastic function qualitatively demonstrates a dependence on direction. These results suggest that the brainstem is an anisotropic material.
Technical Paper
Lawrence E. Thibault, Thomas A. Gennarelli
Abstract This report discusses the development of brain injury tolerance criteria based on the study of three model systems: the primate, inanimate physical surrogates, and isolated tissue elements. Although we are equally concerned with the neural and neurovascular tissue components of the brain, the report will focus on the former and, in particular, the axonal elements. Under conditions of distributed, impulsive, angularacceleration loading, the primate model exhibits a pathophysiological response ranging from mild cerebral concussion to massive, diffuse white matter damage with prolonged coma. When physical models are subjected to identical loading conditions it becomes possible to map the displacements and calculate the associated strains and stresses within the field simulating the brain. Correlating these experimental models leads to predictive levels of tissue element deformation that may be considered as a threshold for specific mechanisms of injury.
Technical Paper
Thomas A. Gennarelli, Jacob M. Abel, Hume Adams, David Graham
Frontal and temporal lobe contusions that were caused by a single sagittal plane angular acceleration impulse were analyzed. At neuropathological exam the depth, extent, and location of contusions were mapped and described according to a classification previously developed for human use. Of 30 rhesus monkeys subjected to a single angular acceleration impulse, 13 had no frontal or temporal contusion (Group 1), 8 had only frontal contusion (Group 2) and 9 had temporal contusions (Group 3). Correlation with angular acceleration, tangential acceleration and tangential force showed that the three groups were statistically different. The mean peak positive tangential force for Groups 1-3 was 541, 659 and 766 newtons respectively (p<0.10). This suggested that as mechanical imput increased, frontal contusions occur before temporal contusions and that the threshold for frontal contusion is less than that for temporal contusion.
Technical Paper
Britton Harris
An attempt is made to show that a transportation system as such is too "open" for sound planning, and that the larger interaction system (including the users of transportation and their bases at home and work) must be considered. This requires a detailed understanding of micro-behavior, social goals, and public policy alternatives in a wide range of situations.
Technical Paper
A. B. O. Soboyejo
The principle of maximum entropy is used to obtain the prior probability distribution functions for critical creep-strain and creep-rupture characteristics of engineering materials, operating at known high temperatures and uniaxial stresses. From the prior distribution function obtained, reliability function which is simply the probability of successful operation of the material, can be derived for specified critical creep-strain and creep-rupture modes of failure. An attempt is made to derive the reliability functions from prior considerations of the mechanics of failure, and the mechanical and physical characteristics of engineering materials. This work assumes that mechanical creep design reliability functions for creep-rupture and critical creep-strain modes of structural elements can have values such that the failure of the elements can occur either by any of the modes of failure or by the assumed combined modes of failure.
Viewing 1 to 29 of 29


    • Range:
    • Year: