Imaginate

Important results from the project

The project has investigated the possibility of developing a machine learning (ML) model with data from a data logging platform integrated in a car and then integrating the model in the same data logging platform to log the results. The goal was to show that this was possible and the results from the project show that this is the case.

Expected long term effects

The results show that it is possible to train a model with data from an existing data logging platform and then integrate, drive and log the results from the model on the same platform in the car. Machine learning can be used "on the edge" to identify complex events based on information from different sensors. An expected effect of this is that data logging and data annotation can be streamlined.

Approach and implementation

The project has been carried out in different stages: planning, data collection, training of ML models, integration and testing. In summary, the implementation has worked well. The same people have been involved throughout the project and the work between the three project parties has been synchronized in weekly meetings, with close contact during the work.