Znak Politechniki Warszawskiej

Scientists against accidents and dangers on the road

Source: project InDeV

Source: project InDeV

Business - Innovations – Technology

Improved safety of vulnerable road users (such as pedestrians, cyclists, motorcyclists and moped riders) is the main objective of an international project called InDeV (“In-Depth understanding of accident causation for Vulnerable Road Users”) funded by the European Union. Nine universities and research institutes were involved in the undertaking, including the Warsaw University of Technology.

A team of WUT scientists worked under the direction of Piotr Olszewski, DSc Eng., a university professor at the Faculty of Civil Engineering. The team was primarily responsible for conducting a review of study methods applied to understand road user safety issues for vulnerable road users (VRUs). “We have worked through the strengths and weaknesses of each method to prepare recommendations for improvements to safety assessment methods,” says Professor Olszewski. “Our list of references included over 1000 papers.”

In want of precision and cohesion

Drawing on their review, the scientists came up with four key conclusions. The first one is that the number of accidents involving VRUs tends to be underestimated (the Police databases record the number of fatalities but details of injured victims are missing) and this leads to the second conclusion that there is a need to align the Police databases with medical ones (so that accurate information of the actual number of accidents and injured victims could be collected). In their third insight, the scientists point to the fact that the accepted definition of a road accident is insufficient; it captures only events involving vehicles and excludes those where only pedestrians are involved, which prevents accurate identification of all dangers to VRUs. There is no precise description of what makes a road accident injury.

“Various EU countries have different systems of road accident reporting,” explains Professor Olszewski. “There is no cohesion here. For instance, Sweden has a system which aligns the Police database with the medical database so that they will have the total number of accidents rather than just what is reported to the Police. Statistics show a very high number of accidents per million of people in Germany. But this is because they are very precise in counting all accidents, including very minor ones which do not require medical intervention.”

Better prevention

Before proceeding to collect and structure information on methods of traffic safety analysis, the scientists of the Warsaw University of Technology were involved in certain other works.

One of the deliverables under the InDeV project is a handbook entitled: “How to analyse accident causation? A handbook with focus on vulnerable road users.” WUT researchers contributed one of its chapters. “We engaged in consultations with the Motor Transport Institute (ITS), Warsaw Bureau for Mobility Policy and Transport (BPMiT) and Warsaw Municipal Road Administration (ZDM),” says Professor Olszewski. “This book is mainly addressed to practitioners or people who work at offices dealing with transport and mobility.”

One method to get an insight into VRU safety is through traffic conflict observations. “This involves traffic observation and identification of dangerous situations which may lead to an accident,” explains Professor Olszewski.

This means that you do not need to wait for some injury to happen to identify the risks and understand how to prevent them.

This was the method used under the InDeV project. Based on statistics, 27 junctions across 6 countries were selected and the traffic conditions in those junctions were recorded by 2 to 3 cameras. Three locations were recorded for a year and 24 others over a period of three weeks. “In Warsaw, we filmed three junctions: two in the Gocław District (over three weeks in October 2016) and one in the Mokotów District, i.e. the crossroads of Wałbrzyska Street and Harcerzy Rzeczpospolitej Street (from June 26, 2017 through June 13, 2018),” says Professor Olszewski. “It was made possible by our collaboration with the Warsaw Municipal Road Administration team and our colleagues at the WUT Faculty of Engineering, who helped us with some technical aspects, e.g., by developing a method for deep-leaning enabled image recognition.”

Overall, 8 faculty members of the Warsaw University of Technology were involved in work on the project: Piotr Olszewski DSc Eng., University Professor (Faculty of Civil Engineering), Witold Czajewski, PhD Eng. (Faculty of Electric Engineering), Piotr Szagała, PhD Eng. (Faculty of Civil Engineering), Paweł Dąbkowski, MSc Eng. (Faculty of Civil Engineering), Grzegorz Kurzejamski, MSc Eng. (Faculty of Electric Engineering), Beata Osińska, MSc Eng. (Faculty of Civil Engineering), Paweł Włodarek, MSc Eng. (Faculty of Civil Engineering) and Anna Zielińska, MSc Eng. (Faculty of Civil Engineering).

Although the InDeV project has formally closed, the scientists will continue to build on the accumulated experience for the future. “Next year, we are holding an international conference on methods of vulnerable road user safety analysis,” says Professor Olszewski.

An international team

The InDeV project comprised seven work packages in total. These work packages included: a review of study methods for VRU safety analysis; observational field studies at selected sites; development of tools for automated data collection and analysis; a socio-economic cost analysis for costs of accidents involving vulnerable road users; and compilation of a VRU safety analysis toolbox. The project was complete with a package dedicated to the management of all the work and, what is essential at this stage, dissemination and exploitation of the results obtained.

The InDeV project is a joint effort of scientists from Sweden, Denmark, Belgium, Germany, the Netherlands, Spain, Norway, Poland and Canada.

Agnieszka Kapela

Promotion and Information Office

Learn more about the project at https://www.indev-project.eu/InDeV/EN/Home/home_node.html

Project duration: May 1, 2015 to October 31, 2018.

This project is funded by the European Union’s Horizon 2020 Framework Program under grant agreement No. 635895.