Reconstruction
Crash Tests
Experts
Research
Database
Congresses

To prove the possibility to park the Ferrari F430 in the garage was necessary to check the possibilities of entering the garage by the car. The request from the court was to investigate several variants of the entry option using 3D simulation. In different directions were necessary determine the contact of the vehicle with the road or the garage.

In the 2009 the area was measured by GPS total station. Because of the lack of sufficient number of points, the area was captured with laser scanner in 2017. At the time of scanning a sufficient number of photos were created. Photos were used to build a point cloud using the Agisoft. The point clouds were used to comprehensively assessment of the insurance fraud. Three different polygons made from point clouds were compared in PC Crash.

(EVU-members can download the full article)

Data recorded by vehicle-mounted electronic control units (ECUs) are highly useful for reconstructing traffic accidents. In this context, event data recorders (EDRs) are airbag ECU components used to log information from crash events for a period of around 5 s before a collision. Information about accelerator/brake-pedal operation is very useful for understanding pre-crash driver actions, but the accuracy of EDR speed data must be checked in this regard. Such data are unlikely to reflect the actual speed during brake-related skidding, for example, because they are determined from the rotational speed of the drive train. Therefore, it is important to check the accuracy of EDR speed data in accident reconstruction. Pre-crash safety systems (PCSs) automatically apply braking in hazardous situations, and the relevant data recorded in the ECU are promising in the field of accident reconstruction. This study involved physical tests in the form of simulated rear-end collisions and the collection of recorded PCS data at the post-test stage. A Toyota Corolla was used as the test vehicle and equipped with a Toyota Safety Sense C PCS, which provided single-frame data from the point at which system braking was applied, with information including vehicle speed and driver actions. The speed data produced were highly accurate, having been collected before hard braking was applied. Checking to determine the influence of PCS automatic braking on EDR speed data showed that high accuracy was maintained. However, the EDR data provided no data about automatic PCS braking. Thus, PCS activation cannot be reliabily detected from an accident scene even if PCS activation data are present in the ECU of an antilock braking system.

(EVU-members can download the full article)

Different products of different manufacturers where installed in 5 vehicles. In some of the vehicle two different systems were installed. All the vehicles were tested on the same road track. Each vehicle covered the track 3 times at different hours of the day (including at night and in rain). To determine the ability of the systems to detect the actual road signs (not via a GPS map), some of the road signs were covered with black plastic foils during some of the tests. The system’s alert the driver for each road sign meanings was recorded and evaluated. For comparison an addition of 2 vehicles with GPS system were tested.

The results show the weaknesses and strengths of each system. The research demonstrates when the alerts to the driver are improper. In some cases the driver is alerted in vain, even though no threat is present. In other cases, the system does not alert the driver to dangers lurking “over the horizon”, i.e. not in the system’s field of vision.

(EVU-members can download the full article)

The characteristic of the paint of metal and plastic parts of cars are several cross-section layers which differ according to their colour and chemical composition. The distribution, colour and chemical composition of the car paint layers differ depending not only on the car manufacturer, but also depending on the make, model and year of production of the car of the same manufacturer. This paper presents how the make, model and year of production can be identified by comparing the IR spectra of each layer of an unknown paint sample to the currently available data of different car manufacturers. An unknown paint sample was found on clothes of a pedestrian who suffered an injury in a traffic accident including an unknown car which left the crime scene. The physical characteristics of the unknown paint sample were examined under microscope to identify the distribution, number and colour of the cross-section layers. The qualitative chemical composition of each paint layer was examined by the IR- spectroscopy method and the identified IR spectra were compared to the IR spectra of the car paint from the available ENFSI EUCAP/FRCAP bases using the KnowItAll software.

(EVU-members can download the full article)

The assumed reaction time of drivers, who react in a certain traffic situation, has an elementary importance on the jurisdiction and accident reconstruction. As reaction time can be understood as a period between the moment of the first visibility of a relevant stimulus and the initiation of an appropriate action, different cognitive factors, also known as information processing systems, play a role for the driver. The current contribution discusses, how strong those information processing systems can be influenced through specific situations during road traffic, and thus also through decision making processes. Based on a review study on existing incidence data it will be critically discussed if and how it is reasonable specifying general reaction times in road traffic law. Simultaneously, an empirical method is presented which shall provide assumptions regarding the expected case-specific behaviour of a representative testing sample.

(EVU-members can download the full article)

The paper demonstrates a practical and easy method of producing a 3D point cloud using the Agisoft photogrammetry program, to generate a base plan for the analysis of CCTV footage.

An explanation of how to view CCTV footage as individual frames in Avidemux is explained, so that a vehicle’s position in time and space can be determined.

A point cloud can be imported into PC Crash and the camera position optimised, and thereafter a 3D model of the vehicle introduced and synchronised with the actual position of the vehicle in the CCTV footage.

Using 2 or more frames and defining the position of the vehicle in the 2D window, it is possible to determine an approximation of the distance covered in time and hence the vehicle’s speed by reference to the 3D model in the 3D window.

Using this procedure, based on Freeware or inexpensive Software, the accident investigator can quickly and accurately estimate the vehicle speed from CCTV footage.

 


(EVU-members can download the full article)

Advances in traffic signal data monitoring systems are providing engineers with a big-data solution for designing, monitoring, and optimizing traffic signals. In its most modern applications, a traffic signal’s exact sequence of events prior to and during a collision will be recorded with granular detail and accurate time-stamp and positional information. Meanwhile, a wide variety of GPS-compatible systems are in common use in automobiles, commercial vehicles, and even pedestrian and bicyclist applications. Devices often record coordinate positional data, instantaneous speed, and time-stamp information, which can be combined with the traffic signal data logger records to establish the specific movements of involved parties through space and time. The combination of these two data technologies is applicable to answering common questions involving signalized intersection collisions, including which driver had a green signal, or how the incident was affected by the user responses to signal change intervals, phasing, and pedestrian signal operations.

(EVU-members can download the full article)

The article gives a current overview of the experience of AZT with the complexity and limitations of storing accident and event data and with technical access. It is apparent that a variety of event data is already stored in modern vehicles, but at the same time, a universal standard for all models, brands or manufacturers has not been established. In crash tests, there were good correlations with respect to the CDR data measured by external measuring technology and the CDR data logged by the vehicle – provided that the limitations in data generation in the vehicle were recognized and taken into account. With highly automated vehicles in mind, in the revision of the German Road Traffic Act (§ 63a StVG), the Federal Republic of Germany has made it mandatory to provide standardized and regulated data storage which is restricted to information for determining the driving mode. Regulation of access to this data and the specific configuration of a data model are yet to be provided. Allianz supports the model of an independent data trustee, allowing authorized persons easy, tamper-proof, fair and non-discriminatory access to the relevant data elements.In the work group AHEAD, with the support of AZT and additional cooperation partners, a data model that is intended to allow proper investigation of accidents and events involving highly automated vehicles in the future is being developed. Relevant data elements from the driving data, the environment and object recognition, the driver activity and the crash are to be defined, and data and formats are to be standardized. In crash tests, the designed data model is validated and evaluated. The aim is for this data model to ultimately be able to be used as a basis in the standardization work for the homologation of highly automated vehicles (Level 3 and above).

(EVU-members can download the full article)

Determining the approach and the collision speed are main components of reconstructing accidents. Conventionally, these speeds are determined by impulse, energy and damage considerations.

Due to the increasing amount of recorded data in a vehicle, stored speeds can sometimes be recovered from ECU data after an accident. One tempting way of determining the speed would be equating the approach speed with stored data from error codes and freeze frames.

This approach will be critically assessed in the following article.

(EVU-members can download the full article)

Many of today’s BMW vehicles have a central fault memory that is likely to store speed in accident situations. In this paper we examine the central fault memory, we explore the meaning of the stored speed values and we present a best practice for reading out the memory.









(EVU-members can download the full article)