Sonstige
Filtern
Dokumenttyp
- Konferenzveröffentlichung (42) (entfernen)
Sprache
- Englisch (42) (entfernen)
Schlagworte
- Datenbank (42) (entfernen)
Institut
In 2012 the fifth ESAR conference (Expert Symposium on Accident Research) was held in Hannover. ESAR is an international convention of experts, who analyze traffic accidents all over the world and discuss their results in this context, conducted at the Medizinische Hochschule Hannover every 2 years. It connected representatives of public authorities, engineers in automotive development and scientists and offers a forum with particular emphasis on In-Depth-Analyses of accident statistics and accident analyses. Special focus is placed on research on the basis of so-called "In-Depth-Accident-Investigations" [data collections at the sites of the accidents], which are characterized by extensive documentations of the sites of the accidents, of the vehicles as well as of the injuries, encompassing several scientific fields. ESAR aims at a multi-disciplinary compilation of scientific results and at discussing them on an international, scientific level. It is thus a scientific colloquium and a platform for exchanging information for all accident researchers. Experiences in accident prevention as well as in the complex field of accident reconstruction are stated and new research fields are added. Existing results of long-term research work in Europe, the US, Australia and Japan include different infrastructural correlations and give findings on population, vehicle population and driver characteristics, which offer a basis for recommendations to be derived and measures for increasing road safety.
In order to enable foreseeing or comparing the benefit of safety systems or driver assistance systems in Germany, in the United States and in Japan, the traffic accident databases in those three countries are examined. The variables used are culpable party, collision partner, accident type, and injury level and the method to re-classify the databases for comparison are proposed. The result indicates that single passenger car fatality is the most frequent in Germany and in the United States, while passenger car vs. pedestrian is the most frequent fatality scenario in Japan. When the casualty by fatality ratio is focused, the greatest difference is observed in rear-end collisions. The ratio of slight injuries in Japan yields about eighteen times as many as those in Germany, and about eight times as many as those in the United States.
In order to improve the protection of children transported in cars, within the CHILD programme (GR3D-CT2002-00791) real world road accidents are thoroughly analysed and then reconstructed in laboratory. Prior to comparing injury severities of real victims to physical parameter values measured on the dummies, the quality of the reconstructions is evaluated by experts who use their experience based on the investigation of numerous and various accidents. This paper presents a new tool aiming at better evaluating and validating accident reconstructions. It is based on statistical evaluation of vehicle deformations which gives weighing factors for every part of the car body structure finally leading to a specific Reconstruction Quality Score (RQS indicator). Furthermore, the reliability of this score, depending on the number of measured points, can be established. This tool includes a function aiming at adjusting the speed for a further reconstruction and at defining the launching speed and the pulse shape for complementary sled tests. Finally, the functions of the RQS software and database are presented.
While it is important to track trends in the number of road accidents in different countries using national statistics, there is a need for data with more detailed information, so called in-depth accident data. For this reason, several accident data projects emerged worldwide in recent years. However, also different data standards were established and so comparative analysis of international in-depth data has been very hard to conduct, so far. This is why the project iGLAD (Initiative for the Global Harmonization of Accident Data) was established and created the prerequisites for building up a standardized dataset out of the common denominator of different in-depth accident databases from Europe, USA and Asia. In the first phase, the project received funding from ACEA to compile an initial database. To accomplish this, a suitable data scheme has been defined, a pilot study has been conducted as proof of concept and the recoding of the first common data base has been initiated. Also, to prepare the project for its self-supporting continuation in the next years, a business model has been developed. This paper reports the history and status of the project, the current challenges and the creation of a capable consortium to maintain the data. In mid-2014, the initial database containing 1550 cases from 10 different countries will be completed and a first detailed view on this data will be possible.
Police records about traffic accidents like used by IRTAD (International Road Traffic and Accident Database) and CARE (Community Road Accident Database) do not represent all road injuries. For instance, road accidents of bicyclists without a counterpart are usually not reported. Furthermore, IRTAD-like data contains hardly any information on injury outcome and accident circumstances. This information gap leads to an under-representation of the safety concerns of the most vulnerable road users like children and the elderly both in accident research and safety promotion. Injury registration for the European Injury Database (IDB), in turn, combines details of accident causation with diagnostic information that can be used to assess injury severity and long term consequences. The IDB is collecting data from hospital emergency department patients and is being implemented in a growing number of countries. In this article IDB results on mode of transport and injury outcome are presented from a sample of nine EU member states.
In Germany averagely two million traffic accidents happen each year and emergency medical services are called to more than 400 000 patients. Even though this number is decreasing continuously (due to improvements in the fields of vehicle safety, road construction, and accident prevention) every case is yet a challenge for the rescuers and requires improvements in emergency medicine as well. Especially during diagnostics right at the accident scene, there are only limited instruments available to gain the necessary knowledge of the injuries suffered, to come to essential decisions about treatment or transport. To provide an additional diagnostic aid by scouting and estimating the situation, a software-tool calculating the likeliness of the most frequent severe injuries (AIS 3-6) of front occupants in passenger cars has been developed to deliver this necessary information about particular accident scenarios. To achieve this, logistic likelihood functions have been calculated in a multivariate regression analysis analysing all AIS 3+ injuries in the GIDAS database of the years 1999-2006 that happened more than four times
An eCall device has been mounted on some vehicles in France since 2003. It is an integrated car radio/GSM/GPS system that can be used with a SIM card. When an accident occurs, a call can be sent manually or automatically made to a telephone call centre. Knowing the geographic location, the vehicle identity and the possibility of a direct communication with the people involved enables the nearest emergency services to be called out. In this context, the LAB / CEESAR have set up a study aimed at evaluating the effectiveness of this system. The purpose of this paper is to detail the E-call system evaluation method of effectiveness used and give a global synthesis of the results.
Impact severity is a fundamental measure for all in-depth crash investigation projects. One methodology used in the UK is based on the US Calspan software package CRASH3. The UK- in-depth crash investigation studies routinely use AiDamage3 a software package which is based on an updated version of the original CRASH3 algorithm, including enhancements to the vehicle stiffness coefficients. Real world accident-damaged vehicles are measured and their crush is correlated with a library of stiffness coefficients. These measurements are then used, along with other parameters, to calculate the crash energy and equivalent changes of velocity of the vehicles (delta-v), which is a measure of the impact severity. UK in-depth accident studies routinely validate the crash severity methodologies applied as the vehicle fleet changes. This is achieved by analysing crash test data and using the appropriate residual crush damage and other inputs to AiDamage3 and checking the program- outputs with the known crash severity parameters. This procedure checks, at least in part, the default stiffness values in the data libraries and the reconstruction methods used.
The role of a national motor vehicle crash causation study-style data set in rollover data analysis
(2010)
On 1 January 2005, The National Highway Traffic Safety Administration, an agency of the United States Department of Transportation, implemented a new data collection strategy designed to assess crash avoidance technologies and report associated behavioral inputs and outcomes. The original goal was a six-year program, however, during the shortened data collection period; it proved a valuable resource for understanding a precrash environment previously obscured by forensic case investigation. Another unintended consequence was an overlap with infrastructure, roadway geometry, and design with the occupant and vehicle outcomes, by virtue of well-defined attributes. External to the collected data, supplementary information was extrapolated, by using manuals published in the United States, by the American Association of State Highway Transportation Officials and selected State Departments of Transportation, in conjunction with the National Motor Vehicle Crash Causation Study (NMVCCS). This provided a backdrop to the infrastructure framework of the rollover problem within which the occupant and vehicle outcomes were studied. If a NMVCCS-style data collection were to be implemented elsewhere, then complementary manuals produced by federal transportation officials might be consulted producing similar relationships. The current study uses NMVCCS data to describe vehicles travelling through diverse design geometries and the outcome for occupants involved in crashes within that system. Codified and extrapolated data form the basis for assessing NMVCCS and its value to the transportation safety community, as the protocols are applicable universally. The benefit in continuing a NMVCCS-style study is noted, as the interaction of roadway infrastructure and occupant protection agencies might find paths to better work together in solving the complex rollover problem using a common data-driven approach.
Road condition acquisition and assessment are the key to guarantee their permanent availability. In order to maintain a country's whole road network, millions of high-resolution images have to be analyzed annually. Currently, this requires cost and time excessive manual labor. We aim to automate this process to a high degree by applying deep neural networks. Such networks need a lot of data to be trained successfully, which are not publicly available at the moment. In this paper, we present the GAPs dataset, which is the first freely available pavement distress dataset of a size, large enough to train high-performing deep neural networks. It provides high quality images, recorded by a standardized process fulfilling German federal regulations, and detailed distress annotations. For the first time, this enables a fair comparison of research in this field. Furthermore, we present a first evaluation of the state of the art in pavement distress detection and an analysis of the effectiveness of state of the art regularization techniques on this dataset.