Skip to main content

Sustainable Connected Cars Pilot

Sustainable Connected Cars Pilot

The Sustainable Connected Cars pilot will be carried out by SoFLEET, Autoaid and Answare. This pilot is focused on cars that belong to different sort of companies interested in achieving an efficient management of their fleets. Thanks to OBD (On Board Diagnostics) dongle devices installed on each car and a Big Data architecture, the partners will apply techniques and algorithms to offer a decision support system for preventive maintenance of fleets, monitoring and promotion of eco-friendly driver behaviours and identification of traffic congestions. These dongles gather valuable information from the vehicle such as location, speed, accelerations and fault codes (DTCs). The Big Data infrastructure will gather this huge amount of information coming from every OBD dongle to analyse car positions, fuel consumptions, engine status, speeds, temperatures, accidents identification, strong accelerations and decelerations, and other parameters. The results will enhance management of car fleets and could provide data to external services or stakeholders such as smart highways, urban transport and insurance companies.

Sustainable Connected Cars Pilot

Fault code messages from electronic control units and drivers feedback will be used to obtain breakdown predictions. Fault codes interpretation is hard to achieve and it differs among vehicle brands and models. However, Autoaid will build a comprehensive diagnostic data base where these codes will be interpreted and classified by severity.

SoFLEET will provide access to the connected car fleets and their associated data. SoFLEET will have to set up an IT infrastructure to anonymize and publish data coming from cars. Additionally, it will provide a visualization interface for data analysis.

Answare will be in charge of coordinating the “sustainable connected cars” pilot. Furthermore, it will implement the Big Data infrastructure, which will make use of Machine Learning and pattern recognition algorithms for breakdowns prediction, reduction of fuel consumptions and identification of traffic congestion areas.