Due to the increasing volumes of bicycle traffic over the years, especially in inner-city areas, as well as the increase in sales of cargo bicycles, there is an increasing need to adapt infrastructures to the new needs. The dimensions of the "new" bicycle categories alone indicate different usage demands on the infrastructure and an increasingly wide spread of riding behavior. The basic prerequisite for use-oriented infrastructure planning that takes into account bicycle composition is the collection of reliable count data and knowledge about the riding behavior of cyclists with different bicycles. At present, it is not (yet) possible to differentiate between different types of bicycles when recording bicycle traffic. In traffic planning, automatic bicycle permanent counting stations are used to enable a comprehensive and permanent recording of crossings over defined cross-sections. In addition to more specialized solutions, there are basically three technology approaches on which automated counting stations are based: Radar counters, camera systems and counters with induction loops. Induction loop technology in permanent counting stations for bicycle traffic is very common. Using induction loops, it is possible to record the majority of bicycle traffic as well as the direction and a classification of speed. A limitation is currently the differentiation between different classes of bicycles, as the currently used counting technology cannot detect whether a conventional bicycle, a cargo bicycle, or a bicycle with a trailer has passed the induction loops. The NUErLast project aims to fill this gap by answering the following questions:

1. What categorization of bicycles is appropriate for further use of the data by municipalities and public works authorities?

2. Which types of cargo bicycles can be distinguished by means of induction loops?

3. Which boundary conditions influence the accuracy of the classification?

4. Which basic statements can be derived from the composition of different bicycle types with regard to the flow of bicycle traffic?

At the beginning, use cases for the application of differentiated bicycle traffic counts are identified in expert workshops. Induction loop data and video data will be collected to identify the electromagnetic signatures of cargo bicycles on the one hand and to derive statements on bicycle traffic flow theory on the other hand. The results will be used in the development of methods for automatic differentiation of bicycle cubatures and thus their classification. Based on the video data, parameters for the bicycle traffic flow theory will be derived.

• Testing the feasibility of classifying detection and counting of cargo bikes using induction loops.
   
• Identification of use cases of differentiated bicycle traffic counts in infrastructure planning and traffic engineering.
   
• Implementation of the gained knowledge into the cycling flow theory.

BMDV - Federal Ministry for Digital and Transport

01 March 2023 until 29 February 2024

Univ.-Prof. Dr.-Ing. Heather Kaths
kaths[at]uni-wuppertal.de
+49 (0)202 / 439 4433

M. Sc. Aboozar Roosta
roosta[at]uni-wuppertal.de
 

The project is being worked on together with

• DTV Verkehrsconsult GmbH, Aachen and
• EcoCounter GmbH, Köln