1.1:  Predicting driver risk: Examine the CanDrive dataset to see what parts of the data are useful.

The CanDrive study collected in-car data of older drivers circa 2015. This unique dataset includes information on the drivers’ cognitive, physical, and general health aspects. Additionally, this dataset is an opportunity to include data from semi-autonomous vehicle systems.

1.2: Predicting driver risk: Examine the CAA dataset to see what parts of the data are useful.

The CAA has also collected in-car data. Their dataset however only includes drivers insured by the CAA. The outcome includes historical and current in-car information from over 30,000 drivers of all ages. This dataset does not provide information about the drivers’ health. However, it does have basic age and gender information. Furthermore, the dataset shows what vehicle sensors in both older and newer cars can capture.

1.3: Consult with human factors experts, application developers, the NRC life expert panel, and representatives of the Driver Safety Team (DST): Their expertise will help validate the data that we selected from the CanDrive and the CAA datasets.

Not all elements contained in the CanDrive and the CAA datasets are relevant. For that reason, the project team and academic collaborators will explore which elements are important for the SENSE-MD project. Then, knowledge users will verify the selected data fields. To ensure that the selected data fields do not put any driver group at a disadvantage, the project team, academic collaborators, and knowledge users will consider the data fields from an inclusive (GBA+ and EDI) perspective.

1.4: Summary report on relevant data fields from in-car sensors for assessment of driver risk.

Finally, this report will identify the most relevant part of the data collected from in-car sensors. This will include data from semi-autonomous systems that could be used in the AI system. Key activities will be the identification of measures that can be used for an assessment of the driver’s action/in-action. Knowledge users will provide feedback on the summary report.