Documentation:FIB book/Head Injury in Vehicle-Pedestrian Collisions

From UBC Wiki

Head injuries are the most common type of injury in pedestrian-vehicle collisions, and are often serious or life threatening.[1] Between 2006 and 2015 in the U.S., the annual rate of fatalities in pedestrian-vehicle collisions was 1.55/100,000 population, and the annual rate for injuries in pedestrian-vehicle collisions was 21.89/100,000 population[2]. Elderly populations had the highest fatality rates, while people aged 15-19 had the highest injury rate in the population.[2] There are two main causes of severe head injury in a pedestrian-vehicle collision. A primary impact occurs when the pedestrian’s head strikes the hood of the vehicle. A secondary impact occurs if the pedestrian’s head impacts the ground after the initial collision[3]. Head Injury Criterion (HIC) can be used to quantify the likelihood of severe injury in both primary and secondary impact scenarios.

Modelling Vehicle-Pedestrian Collisions

Anthropometric Test Devices

Full body Anthropomorphic Test Devices (ATDs) are not used in regulatory pedestrian impact tests. Instead, sub-system ATDs for head, lower leg and upper leg are used exclusively.[4] The three dummy parts are propelled toward a stationary test vehicle, at the specified testing speed; in reverse from a real world accident scenario with a relatively stationary pedestrian and moving vehicle.

Pedestrian impact tests with ATDs are performed to ensure new vehicles fall within acceptable design parameters and provide a low risk of injury or death to the general public. For the head, a HIC value is found. One area of concern is the use of the single standardized impact speed when performing tests. HIC has been found to vary with impact speed. By assessing vehicles at only one impact speed, pedestrian safety from a vehicle may be overestimated.[5]

Linear Spring Model

The following linear spring model relating impact speed to HIC for a given test location was developed by Searson, Anderson & Hutchinson.

[5]

When utilizing the linear spring model, a reference HIC value with corresponding reference speed is found from published test data. The subscript "1" corresponds to a reference HIC and corresponding reference velocity, typically found by looking up New Car Assessment Program (NCAP) values. The subscript "2" corresponds to the unknown HIC or velocity.

Industry standard testing relies on values from testing at one impact speed (40 km/h EuroNCAP, 35 km/h Global Technical Regulation on pedestrian safety (GTR) and Japanese NCAP).[5] Controversy exists about this practice, and the representativeness of real world driving conditions. Usage of this model may allow for better pedestrian safety prediction across the range of normal operating speeds (15-50 km/h). The model breaks down for speeds below 15 km/h and above 50 km/h[5]. Further model development is required for prediction of HIC for highway and above-speed-limit city driving.

Computer Simulations

Computer simulations are widely used to model pedestrian-vehicle collision outcomes. They are in general more cost effective, take less time, and can be used to simulate a variety of complex scenarios as compared to cadaver or ATD tests[6]. In the mid 1980s, Niederer, P.F. & Schlumpf, M.R. utilized computational models such as the Calspan Crash Victim Simulation Program alongside ATD and PMHS experiments[7]. However, they determined the accuracy of the technology to be limited in its agreement with PMHS and ATD models during sled tests[8]. The authors also noted that pedestrian impacts are highly variable due to the pedestrian often being in motion prior to the impact. Incidents where a pedestrian was struck by the corner of a vehicle were identified as being significantly more difficult to model due to the complex geometry of the situation. With an increase in computing power, simulations have become more complex and reliable. Models are validated against PMHS biofidelity corridors[9] to ensure accurate material properties and responses. Simulations of both the pedestrian and the vehicle front are generated using finite element or articulated rigid body models, or a combination of the two[6]. Different models can be created or scaled and validated to perform tests with a variety of human and car models, allowing data to be collected on demographics that are not represented in ATD or PMHS tests[9].

When working with data from tests on postmortem human surrogates (PMHS) of diverse body types and combining with pedestrian finite element models (PFEM), generic anthropometric models yield low biofidelity[10]. In these cases, it is necessary to apply global geometric personalization via mesh morphing the PFEM. This scales the model to match the anthropometry of each distinct PMHS. The morphing process is only applied to geometric dimensions, and does not account for variations in local density that may differ between distinct PMHS.

Mesh morphing is of particular recent interest with an increasing proportion of US individuals being elderly or obese individuals.[11] The paper from Poulard, D., Chen, H., & Panzer, M. compared the kinematic behaviour of two obese male PMHS with differing anthropometry, with a baseline version of the AM50 THUMS PFEM and with versions morphed to match each distinct anthropometry. Noticeable kinematic variation was observed from the baseline model, with differing head impact timing and head impact location. This impresses the necessity for inclusion of obese models when performing new vehicle safety testing, as the probability and severity of injury differs from baseline models.

Influence of Vehicle Factors

Vehicle Speed

Few studies exist that have formally investigated the impact of speed on HIC for struck pedestrians.[5] However, an resulting increased HIC from an increased impact speed can be logically concluded from the known kinematic behaviour of human bodies. From the finite-element model tests performed by Liu et al.,[12] with a headform striking a windshield of known mechanical properties, a clear increase increase in HIC with impact speed was demonstrated. Studies using MADYMO models of a dummy struck by a vehicle have shown the probability of a HIC value greater than 1,000 is more than 50% for an impact speed of 40 km/h, while comparatively less than 25% at 30 km/h.[13]

Further information about the impact of speed on HIC can be found in the previous section on the Linear Spring Model.

Vehicle Geometry

Vehicle geometry has been identified as a significant factor in both primary and secondary head injury scenarios. Niederer and Schlumpf[7] utilized PMHS and ATD sled tests to determine injury risk from various car geometry configurations. They determined that the most significant factors were the height of the upper leading edge of the car, and bumper lead angle. Both a high leading edge and high lead angle tended to decrease the impact speed of the pedestrian’s head onto the hood of the car. Increasing both these factors does increase the load on the pelvis, however the authors determined the overall injury would still be reduced. A limitation of this study was that the optimal geometry for injury reduction is only valid for an adult of average height, which does not account for the variety of of ages and body types who are harmed in pedestrian collisions[2].

Gupta and Yang[3] also discuss the influence of vehicle geometry on pedestrian head impacts, but instead analyzed secondary head impacts during a collision. This study included SUV vehicle profiles in addition to car profiles, which is important to consider as the popularity of SUVs among the general population increases[14]. Using validated finite element vehicle profile models and MADYMO human models as pedestrians, they ran a variety of crash simulations and determined the HIC15 for the impact as well as the overall kinematics. They found that for an SUV, a lowered front profile results in secondary head impact with the ground, while collision with a raised front profile leads to the lower extremities impacting the ground, avoiding severe head injury. Conversely, for a midsize sedan a lowered front profile results in the lower extremities impacting the ground first, while a raised front profile results in head-on secondary impact with the ground and higher HIC values. It is also specified in the paper that at speeds of 40km/h, SUV-pedestrian collisions always resulted in a secondary head impact with the ground regardless of profile. These findings were valid for both a 50th percentile male and 5th percentile female in the simulations, which illustrates an advantage of complex computer simulations over physical testing methods.

These findings indicate a large area of future research. First, it is not clear that these findings have been taken into account by car manufacturers. A critical area of future work will be applying these results to minimize the severity and occurrence of head impact and injury. The second area of future work is to reconcile the apparent trade-off between these two studies. For midsize sedans, Niederer and Schlumpf[7] determined that a higher leading edge and lead angle decreased primary head impact, but Gupta and Yang[3] found that these same properties increased secondary head impact. More work should be done to determine how to reduce overall head injury risk.

Hood Stiffness

Diagram of human head impacting a vehicle hood, causing hood deformation and subsequent impact to substructure. "Bottoming out"

Hood stiffness plays an important role in severity of pedestrian injury. Our analysis of the literature provided surprisingly few specific values when discussing stiffness,[13][1][15] instead comparing the relative stiffness of different car structures. Stiffer structures result in greater severity of injury. By lowering the contact stiffness, both linear acceleration and impact force of the head are reduced.[16] However, lowering stiffness must be balanced with avoiding bottoming out.

Bottoming out occurs when the pedestrian head impacts with sufficient force to deform the hood far enough for collision with a substructure. One common example of bottoming out substructure is the engine block. These substructures typically have a higher stiffness than the hood, leading to increased pedestrian injury.[5] Bottoming out is not accounted for in the linear spring model, even within the 15-50 km/h range. Due to the possibility for higher severity pedestrian injuries, accurate modelling of bottoming out is a future research area for greater pedestrian safety.

Preventative Technology

Autonomous Emergency Braking

Autonomous Emergency Braking (AEB) attempts to reduce impact speed, and thus reduce harm to pedestrians both in number of collisions and severity of resulting struck pedestrian injuries.[4] It uses a camera, laser-ranging, infrared or other sensor system to detect if a collision is imminent. Collision detection data is passed to a computer system, which analyses the probability of collision, current vehicle speed and other environmental parameters. Brakes may be automatically applied to reduce collision severity. Risk equivalence can be performed to compare values of HIC between vehicles with and without AEB systems installed, and for vehicles with different levels of AEB performance. Early simulations with a wide angle system indicate a mean reduction of crash speeds by 25%.[17] In tests performed by Rosén, E. et al., an AEB system demonstrated a mean 43% reduction rate of fatal injuries.

Currently, AEB inclusion in vehicle design is rewarded in the points system of the Australasian NCAP. Both EuroNCAP and GTR are considering similar implementation in their own NCAPs. However, a handful of concerns exist around AEBs. Allowing AEB inclusion to inflate regulation rating may result in less implementation or research into additional safety features, as AEB is comparatively easy and cost effective to implement.[4] Real world AEB systems vary greatly in performance and may not accurately be represented in the risk assessment equation. Controversy about the inclusion of the AEB in the points system exists, as AEB has yet to be proven in the field, and has only been proven in laboratory testing.


References

  1. ↑ 1.0 1.1 Li, G., Wang, F., Otte, D., & Simms, C. (2019). Characteristics of pedestrian head injuries observed from real world collision data. Accident Analysis & Prevention, 129, 362–366. https://doi.org/https://doi.org/10.1016/j.aap.2019.05.007
  2. ↑ 2.0 2.1 2.2 Chong, S.-L., Chiang, L.-W., Allen, J. C., Fleegler, E. W., & Lee, L. K. (2018). Epidemiology of Pedestrian–Motor Vehicle Fatalities and Injuries, 2006–2015. American Journal of Preventive Medicine, 55(1), 98–105. https://doi.org/https://doi.org/10.1016/j.amepre.2018.04.005
  3. ↑ 3.0 3.1 3.2 Gupta, V., & Yang, K. H. (2013, November 11). Effect of Vehicle Front End Profiles Leading to Pedestrian Secondary Head Impact to Ground. SAE Technical Paper 2013-22-0005, https://doi.org/10.4271/2013-22-0005
  4. ↑ 4.0 4.1 4.2 Searson, D. J., Anderson, R. W. G., & Hutchinson, T. P. (2014). Integrated assessment of pedestrian head impact protection in testing secondary safety and autonomous emergency braking. Accident Analysis & Prevention, 63, 1–8. https://doi.org/10.1016/J.AAP.2013.10.014
  5. ↑ 5.0 5.1 5.2 5.3 5.4 5.5 Searson, D. J., Anderson, R. W. G., & Hutchinson, T. P. (2012). The effect of impact speed on the HIC obtained in pedestrian headform tests. International Journal of Crashworthiness, 17(5), 562–570. https://doi.org/10.1080/13588265.2012.699271
  6. ↑ 6.0 6.1 Vychytil, J., Hyncik, L., Manas, J., Pavlata, P., Striegler, R., Moser, T., & Valasek, R. (2016, April 5). Prediction of Injury Risk in Pedestrian Accidents Using Virtual Human Model VIRTHUMAN: Real Case and Parametric Study. SAE Technical Paper 2016-01-1511, 2016 https://doi.org/10.4271/2016-01-1511
  7. ↑ 7.0 7.1 7.2 Niederer, P. F., & Schlumpf, M. R. (1984, October 1). Influence of Vehicle Front Geometry on Impacted Pedestrian Kinematics. SAE Technical Paper 841663, https://doi.org/10.4271/841663
  8. ↑ Niederer, P. F., Schlumpf, M., Mesqui, F., & Hartmann, P.-A.. The Reliability of Anthropometric Test Devices, Cadavers, and Mathematical Models as Pedestrian Surrogates. SAE Technical Paper 830184, 1983, https://doi.org/10.4271/830184
  9. ↑ 9.0 9.1 Vychytil, J., Spicka, J., Hyncik, L., Manas, J., Pavlata, P., Striegler, R., … Valasek, R. (2017). Novel Approach in Vehicle Front-End Modeling for Numerical Analyses of Pedestrian Impact Scenarios. WCXTM 17: SAE World Congress Experience. https://doi.org/https://doi.org/10.4271/2017-01-1451
  10. ↑ Poulard, D., Chen, H., & Panzer, M. (2016, April 5). Geometrical Personalization of Pedestrian Finite Element Models Using Morphing Increases the Biofidelity of Their Impact Kinematics. https://doi.org/10.4271/2016-01-1506
  11. ↑ Hwang, E., Hallman, J., Klein, K., Rupp, J. et al., "Rapid Development of Diverse Human Body Models for Crash Simulations through Mesh Morphing," SAE Technical Paper 2016-01-1491, 2016, https://doi.org/10.4271/2016-01-1491.
  12. ↑ Liu, B., Zhu, M., Sun, Y., Xu, J., Ge, D. and Li, Y. 2010. Investigation of energy absorption characteristics of PVB laminated windshield subject to human head impact. Applied Mechanics & Materials, 34–35: 956–960. https://doi-org.ezproxy.library.ubc.ca/10.4028/www.scientific.net/AMM.34-35.956
  13. ↑ 13.0 13.1 Liu, X. J., Yang , J. K. & Lövsund, P. (2002) A Study of Influences of Vehicle Speed and Front Structure on Pedestrian Impact Responses Using Mathematical Models. Traffic Injury Prevention, 3:1, 31-42. https://doi.org/10.1080/15389580210517
  14. ↑ White, M. J. (2004). The “arms race” on American roads: The effect of sport utility vehicles and pickup trucks on traffic safety. The Journal of Law and Economics, 47(2), 333-355.
  15. ↑ Mizuno, K. and Kajzer, J., "Head Injuries in Vehicle-Pedestrian Impact," SAE Technical Paper 2000-01-0157, 2000, https://doi.org/10.4271/2000-01-0157.
  16. ↑ Yang, J. (2005) `Review of injury biomechanics in car-pedestrian collisions', Int. J. Vehicle Safety, Vol. 1, Nos. 1/2/3, pp.100-117.
  17. ↑ Rosén, E., et al. (2010). Pedestrian injury mitigation by autonomous braking. Accident Analysis & Prevention. 42, 1949-1957. https://doi.org/10.1016/j.aap.2010.05.018.


External Links

Retrieved from "https://wiki.ubc.ca/index.php?title=Documentation:FIB_book/Head_Injury_in_Vehicle-Pedestrian_Collisions&oldid=578521"