To utilize the potential of vehicle-originated sensor data as data source for road operators the CarSense project surveys, tests and assesses (1) different existing and probably upcoming vehicle sensor technologies as potential data sources. The project (2) assesses the technical, economical and data-related value for using the data for road operation and maintenance, (3) proposes distinct recommendations on data categories being appropriate for efficient and beneficial acquisition and usage and (4) proposes a system architecture for seamless integration of vehicle-originated sensor data into the ASFINAG system

landscape.

Being equipped with 80 to 100 sensors, modern vehicles offer enormous potential for different use cases in the context of intelligent transport systems. Commercial services al-ready gain sensor information from vehicles for detecting delays in travel times or adapting routes to the current traffic situation. The goal of the CarSense project is to raise the potential of vehicle-originated sensor information for infrastructure operators.

Effective and efficient operation and maintenance of infrastructure raises a high demand foroperators to continuously collect qualitative and quantitative information on the status of thespatially distributed road infrastructure. Until now, data collection, if automated, is done with road side sensor units as well as controlled test runs of measurement vehicles. Beside high costs for acquisition, operation and maintenance of measurement equipment, measurements are commonly bound to single positions or single time periods. The CarSense project addresses the question, whether measurement data for operation and maintenance can also be efficiently and network-wide acquired from sensors of those vehicles, using the infrastructure on a daily basis. The approach should help to avoid or reduce former draw-backs such as singular measurement positions, additional test runs or additional costs for road side infrastructure acquisition and maintenance.

The project work of CarSense is structured into 4 phases. Phase 1 contains a comprehensive survey on sensor technologies (bottom-up approach) as well as a survey on measurement equipment and requirements of the Austrian highway operator ASFINAG (top-down-approach). Both approaches are used as foundation for a goal- and requirements-driven project design. The survey on sensor technologies includes sensors, which are already in wide-spread use in vehicles, sensors, which can be easily retrofitted into vehicles and sen-sors, which have a high possibility to come to wide-spread use in future vehicles. Together with the results from the requirements study it is possible to identify those sensors, data and tasks, which have the potential to optimize operation and maintenance tasks of road operators.

Phase 2 focuses on the evaluation of at least one sensor technology as foundation for the assessment of sensor technologies in phase 3 for ensuring applicability of the selected technology for a specific use case as well as providing well-founded evaluation results for sensor technology assessment.

Phase 3 pursues the goal of determining the value of vehicle-originated sensor data regarding the potential of efficiently supporting operation and maintenance tasks of road operators from a technical, economical and data-related perspective (primary criteria) as well as from a legal, data privacy as well as social and ethical perspective (secondary criteria). These assessments are used for compiling a comprehensive recommendation catalogue.