Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: traffic study

Sortuj według:

Ogranicz wyniki do:

Estimating the potential of a warning system preventing road accidents at pedestrian crossings

Ess Juri, Luppin Janek, Antov Dago

LogForum

2021

Vol. 17, no. 3

441--452

Background: The safety of pedestrians is one of the main traffic safety issues today and despite measures being applied, the number of pedestrian deaths in traffic is not changing. According to the Pareto Rule, 80% of consequences come from 20% of the causes and here the question arises whether we have already used these 20% of the most efficient measures. Today the European Union (EU) puts big hopes are on contemporary technologies, such as Advanced Emergency Braking Systems (AEB) and cooperative intelligent transport systems (C-ITS). This decade, we can expect smarter vehicles with automatic brakes, and smarter infrastructure which can communicate with vehicles. Along this other profits technological development provides new opportunities for improving pedestrian safety. One of the most promising solutions is deployment of C-ITS systems at uncontrolled crossings. It would monitor the situation and warn the road users of potential dangers as well as make the vehicles brake automatically. However, before making large investments into this field, one has to be sure that this approach will work. The aim of this paper is to describe typical vehicle-pedestrian crash scenarios and to estimate whether a C-ITS warning system is able to prevent them. Research estimates the potential of this system and provides insights to its must-have features. Methods: To understand the situations in which the warning system should function, researchers carried out traffic conflict studies at uncontrolled crossings with traffic filmed in both winter and summer. They determined and described serious conflicts and, based on their scenarios, classified them into three types. Then, researchers selected the most critical conflict of each type and analysed whether warning signals can be provided to the vehicle and the driver early enough to prevent collisions. For these purposes, researchers used a modelling software for traffic accident investigation. To access the efficiency of the C-ITS warning system, researchers estimated the probability of preventing collisions and used the efficiency parameters of classical traffic calming measures. Results: The C-ITS warning system has good potential in preventing vehicle-pedestrian collisions at uncontrolled pedestrian crossings. It is remarkable and very promising that it would be able to prevent all types of conflicts analysed in the scope of this study by warning AEB-equipped vehicles. Warning the driver would be also effective, but the system work will largely depend on the quality of warning signals. An effective C-ITS warning system should be capable of predicting the trajectories and acceleration of road users as well as calculating the stopping distance of vehicles based on the coefficient of static friction. Study showed that in some cases, the system will have to give false positive alarms, but the fewer such alarms will be given, the more efficient the system will be. A disturbing or annoying C-ITS warning system cannot be considered effective. Conclusions: Road accident statistics contain general data about vehicle-pedestrian collisions at uncontrolled crossing, but there is few information about behavioral patterns leading to accidents. Based on large-scaled traffic studies, researchers were able to determine these patterns and described how road users act when being involved in a dangerous situation. This knowledge helped to model typical vehicle-pedestrian collisions as well as their possible scenarios. Researchers used the conflict models totest the C-ITS warning system and to understand its efficiency. The study results were implementedin a prototype that has been developed in Estonia and is being tested it in real traffic conditions of a smart city in the scope of the Finnish-Estonian project “FinEst Twins”. The next steps are to analyze the test results and to conduct research to understand how to warn drivers (and pedestrians) most effectively.

Model systemu monitorowania ruchu drogowego w aspekcie statystycznym

Mitas A. W., Konior A., Bugdol M., Bugdol M., Konior W.

Logistyka

2014

nr 3

4456--4463

W artykule przedstawiono zarys architektury systemu monitorowania ruchu drogowego. Kluczowym atrybutem jest ochrona danych wrażliwych, polegająca na takim rozproszeniu widma informacji, że (potencjalny) nieuprawniony dostęp do danych nie jest równoznaczny z ujawnieniem informacji spersonalizowanej. Zasadniczą cechą systemu jest relatywnie niewielki nakład planowanych kosztów na jego instalację. Monitorowany ruch drogowy jest interpretowany na podstawie niepełnej i nieostrej informacji. Badanie ruchu odbywa się w sposób nieinwazyjny, a sposób identyfikacji przejeżdżających pojazdów nie narusza prywatności użytkowników ruchu drogowego, przy czym poprawność ich poruszania się nie jest celem monitorowania. Podstawą wnioskowania jest informacja niepełna, a odsetek identyfikowanych składników systemu antropotechnicznego, pozostających w ruchu, jest wielowymiarową zmienną losową. W artykule przedstawiono konkretne przykłady z próbnej instalacji, a zasadniczym atrybutem jest oryginalna analiza statystyczna.

The article depicts a scheme of the traffic monitoring architecture. The key attribute is sensitive data protection, which consists of such dispersing the information spectrum that (potential) unauthorised access to them is not equal to revealing personalised data. The essential characteristic of the system is the relatively small costs plan when installing. Monitored traffic is interpreted on the basis of incomplete and blurred information. Traffic study is done in non-invasive manner, and the identification of the passing vehicles does not intrude traffic users privacy, and the correctness of their movement is not the target of monitoring. The basis of inference is incomplete information, and the percentage of identified elements, that are in motion in the the antropotechnical system, is a multidimensional random variable. The article presents particular examples from the test installation, and the essential attribute is the original statistical analysis.