The ASSESS project is a collaborative project that develops test procedures for pre-crash safety systems like Automatic Emergency Braking (AEB). One key criterion for the effectiveness of e.g. AEB is reduction in collision speed compared to baseline scenarios without AEB. The speed reduction for a given system can only be determined in real world tests that will end with a collision. Soft targets that are crashable up to velocities of 80 km/h are state of the art for these assessments, but ordinary balloon cars are usually stationary targets. The ASSESS project goes one step further and defines scenarios with moving targets. These scenarios define vehicle speeds of up to 100 km/h, different collision scenarios and relative collision speeds of up to 80km/h. This paper describes the development of a propulsion system for a soft target that aims to be used with these demanding scenario specifications. The Federal Highway Research Institute- (BASt-) approach to move the target is a self-driving small cart. The cart is controlled either by a driver (open-loop control via remote-control) or by a computer (closed-loop control). Its weight is limited to achieve a good crashability without damages to the test vehicle. To the extent of our knowledge BASt- approach is unique in this field (other carts cannot move at such high velocities or are not crashable). This paper describes in detail the challenges and solutions that were found both for the mechanical construction and the implementation of the control and safety system. One example for the mechanical challenges is e.g. the position of the vehicle- center of gravity (CG). An optimum compromise had to be found between a low CG oriented to the front of the vehicle (good for driveability) and a high CG oriented to the rear of the vehicle (good for crashability). The soft target itself which is also developed within the ASSESS project will not be covered in detail as this is work of a project partner. Publications on this will follow. The paper also shows first test results, describes current limitations and gives an outlook. It is expected that the presented test tools for AEB and other pre-crash safety systems is introduced in the future into consumer testing (NCAP) as well as regulatory testing.
The utilisation of secondary-safety systems to protect occupants has attained a very high level over the past decades. Further improvements are still possible, but increasingly minor progress is only to be had with a high degree of effort. Thus, a key aspect must be the impact to overall safety in an accident. If reliable information is available on an imminent crash, measures already taken in the pre-crash phase can result in a significantly great influence on the outcomes of the crash. With this background preventive measures are the key to a sustainable further reduction of the figures of crash victims on our roads. This paper aims to show a preventive approach that can contribute to lessening the consequences of a crash by creating an optimum interaction of measures in the fields of primary and secondary safety. To further enhance vehicle safety, driver assistant systems are already available that warn the driver of an imminent front-to-rear-end crash. The next step is to support him in his reactions or if he fails to react sufficiently, to even initiate an automatic braking when the crash becomes unavoidable. Automatic pre-crash braking can, in an ideal situation, fully prevent a crash or can significantly reduce the impact speed and thus the impact energy (and the severity of the accident). If a vehicle is being braked in the pre-crash phase, the occupants are already being pre-stressed by the deceleration. The information available about the imminent crash can be used to activate the belt tensioners and likewise other secondary safety systems in the vehicle right before the impact. The pre-crash deceleration also causes the front of the vehicle to dip. Conventional crash tests do not take this specific impact situation into consideration. This is why, for example, the influences of the pre-crash displacements of the occupants are not recorded in the test results. Furthermore, a reproducible representation of the benefit of the vehicle safety systems which prepare the occupants for the imminent impact is not possible. In order to demonstrate the functions of automated pre-crash braking and to investigate the differences during the impact as a consequence of the altered occupant positions as well as the initiation of force and deformations of the vehicle front, DEKRA teamed up with BMW to carry out a joint crash test with the latest BMW 5 series vehicle. It involved the vehicle braking automatically from a starting test speed of 64 km/h (corresponding to the impact speed set by Euro NCAP) down to 40 km/h. The test was still run by the intelligent drive system of the crash test facility. This required several modifications to be made to the test facility as well as to the vehicle. The paper will describe and discuss some relevant results of the crash test. In addition, the possible benefits of such systems will also be considered. The test supplemented the work of the vFSS working group (vFSS stands advanced Forward-looking Safety Systems).
Die UNECE Regelung R58 regelt die Beschaffenheit und die Installation von Heckunterfahrschutzsystemen an schweren Güterkraftfahrzeugen, deren Ziel die Verbesserung der Kompabilität zwischen Pkw-Frontstrukturen und Lkw-Hecks ist. Dennoch verunglücken laut amtlicher Unfallstatistik allein in Deutschland rund 30 Pkw-Insassen in Heckauffahrunfällen auf Lkw tödlich, da diese Vorrichtungen hinsichtlich Einbauhöhe und Steifigkeit den Anforderungen des realen Unfallgeschehens nicht genügen. Das Ziel dieser Studie ist eine quantitative Abschätzung der möglichen Reduzierung der Verletzungsschwere mit Hilfe eines statistischen Modells, die durch eine Anpassung der geltenden Bestimmungen und die damit verbundenen technischen Veränderungen des bereits vorgeschriebenen Heckunterfahrschutzes zu erreichen wäre. In einer Nutzen-Kosten-Analyse wird die Wirtschaftlichkeit dieser Modifizierungen mit einem idealen Notbremsassistenten verglichen. Die Untersuchung orientiert sich dabei an den aktuell in der UN-ECE WP29/GRSG in Genf diskutierten Vorschlägen zur Anpassung der ECE-R58. Das verwendete ordinale Probit-Modell stellt einen Zusammenhang zwischen der Verletzungsschwere im auffahrenden PKW und erklärenden Größen her, in diesem Fall der kinetischen Energie des unterfahrenden Pkws und der strukturellen lnteraktion zwischen Lkw-Heck und Pkw-Front. Diese Maßnahmen könnten demnach 53 - 78% der Getöteten sowie 27 - 49% der Schwerverletzten bei diesen Unfallkonstellationen reduzieren, was pro Jahr 20 Getöteten und 95 Schwerverletzten entsprechen würde. Somit würde eine Modifikation einer bestehenden passiven Schutzmaßnahme an jährlich 100.000 neuzugelassenen Lkw und Anhängern bereits 20 Getötete adressieren. Im Vergleich dazu müssten jährlich 3 Millionen Pkw mit zusätzlicher Sensorik und Aktuatorik für einen idealen Notbremsassistenten ausgestattet werden, um im Idealfall alle Heckauffahrunfälle von Pkw auf andere Pkw oder Lkw und damit 53 Getötete zu vermeiden. Daher fällt auch das Nutzen-Kosten-Verhältnis deutlich zugunsten des verbesserten Heckunterfahrschutzes aus.
Accident research 2.0: New methods for representative evaluation of integral safety in traffic
(2013)
BMW has developed a procedure for rating Advanced Driver Assistance Systems (ADAS) benefits that integrates two distinct tools. The tool "S.A.F.E.R." is designed to analyze the pre-crash phase. The aim of S.A.F.E.R. is to simulate all relevant processes in sufficient detail to obtain reproducible estimates of key indicators (effectiveness, false positives, etc.). The relevant processes include not only traffic and vehicle dynamics, but also environmental and most importantly human factors. Representative distributions of factors and parameters are obtained by taking the stochastic variation of all relevant parameters into account in the simulations. The second tool, known as "ICOS", has been designed to provide a high-resolution, high-fidelity description of crash phase dynamics. If one converts the outputs of stochastic simulation into inputs for crash dynamics, the result is a comprehensive description of exactly how a safety system can reduce injuries. Applications currently focus on high-fidelity simulation of individual crashes in order to enhance our understanding and optimization of connected safety systems. An integrated simulation process thus allows an exact prediction of the effectiveness in individual cases in terms of injury severity. The development and rating of integral safety need to reflect the true efficiency in the field. The integrated approach described here could provide a valid and reproducible basis for rating connected systems of active and passive safety. In particular, "virtual experiments" using a traffic-based approach and incorporating models of all relevant processes constitute an essential element of the approach.
Fahrerassistenzsystemen mit Umfeldwahrnehmung wird ein hohes Potenzial zur Unfallvermeidung zugeschrieben, wenn diese umfassender und intensiver in die Fahrdynamik von Fahrzeugen eingreifen und weiter vernetzt werden. Diese erweiterten Eingriffsmöglichkeiten erzeugen auch neue Risiken, welche vor der Genehmigung und Zulassung für den öffentlichen Straßenverkehr abgesichert werden müssen. Neuartig ist bei diesen Systemen, dass sie nur über eine Situationsrepräsentation die unfallvermeidenden Handlungen ableiten können. Somit kommt zum Risiko des Versagens von Systemkomponenten, das bereits durch die ISO 26262-Norm zur funktionalen Sicherheit adressiert ist, das Risiko aufgrund einer falschen Interpretation auftretenden, nicht situationsgemaessen Auslösung, z. B. durch Situationskonstellationen, die bei der Entwicklung nicht berücksichtigt wurden und daher in den Funktionsspezifikationen nicht enthalten sind. Um die Anforderungen an Absicherungsmethoden für diese Assistenzsysteme zu identifizieren, werden diese zusammengestellt und der Absicherungsaufwand mit bestehenden Methoden, bspw. aufbauend auf den Anforderungen der ISO 26262, bestimmt. Die Analyse zeigt, dass bisherige Ansätze sowohl hinsichtlich der objektiven Nachweisbarkeit der Vollständigkeit der theoretisch möglichen Situationen Lücken aufweisen als auch hinsichtlich des Umfangs der notwendigen Spezifikationen und deren Prüfung in Versuchen. Aufgrund des daher zu erwartenden Aufwands für den Nachweis eines sicheren Verhaltens der Systeme sind eine Priorisierung von Fahrsituationen und die Gewährleistung einer hohen Übertragbarkeit von Bewertungsergebnissen notwendig. Um die Vollständigkeitsproblematik zu adressieren, wird ein Ansatz vorgestellt, der eine objektive Bewertung und den Vergleich von Fahrsituationen ermöglicht. Abschließend werden die Erkenntnisse zusammengefasst und notwendige weitere Schritte für die Schaffung einer einheitlichen Absicherungsstrategie für Fahrerassistenzsysteme abgeleitet.
Abbiegeunfälle mit Kollisionen zwischen rechtsabbiegenden Güterkraftfahrzeugen und Fahrrädern haben in der Regel schwerwiegende Folgen für den ungeschützten Verkehrsteilnehmer. In der Vergangenheit wurde durch eine steigende Anzahl von Spiegeln das individuelle Sichtfeld des Lkw-Fahrers vergrößert und die Sicherheit für ungeschützte Verkehrsteilnehmer durch den Seitenunterfahrschutz verbessert. Da Abbiegeunfälle trotz der Vielzahl an Spiegeln auch heute noch geschehen, gleichzeitig aber Fahrerassistenzsysteme Einzug in viele Fahrzeugklassen gehalten haben, liegt es nahe, derartige Systeme für die Verhinderung von Abbiegeunfällen zu nutzen. Um entsprechende Systementwicklungen fördern zu können oder aber auch Systeme vorschreiben zu können, sind Anforderungen und passende Testmethoden für Abbiegeassistenzsysteme erforderlich. Ziel der BASt war es, solche Anforderungen und ein mögliches Testverfahren hierfür zu entwickeln. Ausgehend von Analysen des Unfallgeschehens wurden charakteristische Parameter und Begleitumstände von Unfällen zwischen Fahrrädern und rechtsabbiegenden Lkw identifiziert. Aus fahrdynamischen Überlegungen folgt bei den gegebenen Parametern, dass nur eine frühe, aber niederschwellige Fahrerinformation eine wirkungsvolle Assistenzfunktion zur Verhinderung der Unfälle sein kann. Für automatische Bremsungen gibt es bisher noch zu wenig Erfahrungen im Feld, und klassische, hochschwellige, aber sehr spät erfolgende Warnsignale würden durch die dann noch verstreichende Reaktionszeit keine rechtzeitige Bremsung des Lkw-Fahrers mehr hervorrufen. Basierend auf dem identifizierten Parameterraum, der zum komfortablen Anhalten erforderlichen Zeit und einem geeigneten Kinematikmodell lassen sich die räumlichen Bereiche um den Lkw definieren, in dem eine Umfelderkennung den Fahrradfahrer detektieren können muss, damit das Informationssignal durch das Assistenzsystem an den Lkw-Fahrer rechtzeitig ausgegeben wird. Aktuell wird davon ausgegangen, dass ein Abbiegeassistenzsystem, das die hier beschriebenen Prüfungen besteht, einen sehr positiven Einfluss auf das Unfallgeschehen zwischen rechtsabbiegenden Lkw und Fahrrädern haben wird.
Euro NCAP will start to test pedestrian Automatic Emergency Braking Systems (AEB) from 2016 on. Test procedures for these tests had been developed by and discussed between the AsPeCSS project and other initiatives (e.g. the AEB group with Thatcham Research from the UK). This paper gives an overview on the development process from the AsPeCSS side, summarizes the current test and assessment procedures as of March 2015 and shows test and assessment results of five cars that had been tested by BASt for AsPeCSS and the respective manufacturer. The test and assessment methodology seems appropriate to rate the performance of different vehicles. The best test result - still one year ahead of the test implementation - is around 80%, while the worst rating result is around 10%. Other vehicles are between these boundaries.
Accidents between right turning trucks and straight riding cyclists often show massive consequences. Accident severity is much higher than in other accidents. The situation is critical especially due to the fact that, in spite of the six mirrors that are mandatory for ensuring a minimum field of sight for the truck drivers, cyclists in some situations cannot be seen or are not seen by the driver. Either the cyclist is overlooked or is in a blind spot area that results from the turning manoeuvre of the truck and its articulation if it is a truck trailer or truck semitrailer combination. At present driver assistance systems are discussed that can support the driver in the turning situation by giving a warning when cyclists are riding parallel to the truck just before or in the turning manoeuvre. Such systems would generally bear a high potential to avoid accidents of right turning trucks and cyclists no matter if they ride on the road or on a parallel bicycle path. However, performance requirements for such turning assist systems or even test procedures do not exist yet. This paper describes the development of a testing method and requirements for turning assist systems for trucks. The starting point of each development of test procedures is an analysis of accident data. A general study of accident figures determines the size of the problem. In-depth accident data is evaluated case by case in order to find out which are representative critical situations. These findings serve to determine characteristic parameters (e.g. boundary conditions, trajectories of truck and cyclist, speeds during the critical situation, impact points). Based on these parameters and technical feasibility by current sensor and actuator technology, representative test scenarios and pass/fail-criteria are defined. The outcome of the study is an overview of the accident situation between right turning trucks and straight driving cyclists in Germany as well as a corresponding test procedure for driver assistance systems that at this first stage will be informing or warning the driver. This test procedure is meant to be the basis for an international discussion on introducing turning assist systems in vehicle regulations.
The project UR:BAN "Cognitive assistance (KA)" aims at developing future assistance systems providing improved performance in complex city traffic. New state-of-the-art panoramic sensor technologies now allow comprehensive monitoring and evaluation of the vehicle environment. In order to improve protection of vulnerable road users such as pedestrians and cyclists, a particular objective of UR:BAN is the evaluation and prediction of their behaviour and actions. The objective of subproject "WER" is development support by providing quantitative estimates of traffic collisions at the very start and predict potential in terms of optimized accident avoidance and reduction of injury severity. For this purpose an integrated computer simulation toolkit is being devised based on real world accidents (GIDAS as well as video documented accidents), allowing the prediction of potential effectiveness and future benefit of assistance systems in this accident scenario. Subsequently, this toolkit may be used for optimizing the design of implemented assistance systems for improved effectiveness.
The evaluation of the expected benefit of active safety systems or even ideas of future systems is challenging because this has to be done prospectively. Beside acceptance, the predicted real-world benefit of active safety systems is one of the most important and interesting measures. Therefore, appropriate methods should be used that meet the requirements concerning representativeness, robustness and accuracy. The paper presents the development of a methodology for the assessment of current and future vehicle safety systems. The variety of systems requires several tools and methods and thus, a common tool box was created. This toolbox consists of different levels, regarding different aspects like data sources, scenarios, representativeness, measures like pre-crash-simulations, automated crash computation, single-case-analyses or driving simulator studies. Finally, the benefit of the system(s) is calculated, e.g. by using injury risk functions; giving the number of avoided/mitigated accidents, the reduction of injured or killed persons or the decrease of economic costs.