Towards Intelligent Traffic Management using anonymized trajectory data for personal multimodal commuting recommendations
Marco Schaarschmidt1, Tim Seidel1, Thorsten Paßfeldt1, Jan Kettler1, Clemens Westerkamp1 *, Thomas Hupperich2, Christian Kray2, Chris Fries2, Caro Niebl2
Abstract
The IIP competence cluster (IIP means intelligent Commuting) combines collected data on the use of different transport to facilitate personal navigation for commuters in a more demand-oriented and resource-efficient way.
Private transport in Osnabrück, Münster and other cities has increased over the years and overloads the roads. To improve this situation, mobility data in the urban area will be recorded and anonymized dynamically, i.e., for specific applications and external solution providers. On an urban data platform (UDP) of the municipal utilities, the existing official and publicly available data, e.g., of the city, are merged. They are supplemented by new data sources from crowdsourcing and citizen participation. The researchers are developing a toolkit for diverse levels of anonymization or depersonalization of personal mobility data with special attention to spatial-temporal referencing. This mobility data can be used for various applications in practice. This enables the cross-data source assignment of objects such as cars, buses, people, cyclists, e-scooters, and others without violating the privacy of road users. The developed modules can be used directly in data sources such as cameras and navigation apps and are made available to the public as open source.
Of course, individual journeys which allow conclusions about single persons and all associated data need to be protected. However, the more the data is anonymized, the more information is lost.
Unlike typical navigation apps, the IIP app will inform commuters to decide about intelligent mobility options as soon as traffic effecting events are foreseen. Many traffic parameters like bad weather conditions, traffic jams and road construction works are known before and can be used to early inform people with commuting routes affected by these events. Commuting alternatives can combine different means of transport and integrate current traffic situations. E.g., cars are directed to a P+R facility and public transport permits reaching the destination easier. To better understand traffic and transport possibilities, a Digital Twin based modelling approach is investigated. Real data are supplemented by synthetic data covering a wider range of possible but rare traffic situations. Dashboards from the University of Münster enable recommendations for action for transport planning.
For personal commuting as well as for the overall traffic situation, the project aims to combine both individual mobility wishes and common good-oriented goals such as the reduction of greenhouse gas emissions and an overall optimization of limited infrastructure resources.
*: Speaker
1: University of Applied Sciences Osnabrück
2: University of Münster