Vehicles are equipped with ever increasingly advanced communication and sensing resources, giving rise to the need for effective mechanisms to utilise and leverage the large amounts of data they gather. For example, this data could be used to predict traffic flow and to monitor driver behaviour.
There are many different, distributed machine learning techniques that could be used to identify patterns and make logical decisions based on the large volumes of data generated. Traditionally, devices readily shared their raw data with a centralised server. However, this sharing of raw data is not conducive to maintaining the privacy of the data. One way that this has been overcome is through the use of Federated Learning. This reduces the volume of data shared, as only the parameters obtained from locally trained models are forwarded to the central server. However, the mobility of vehicles gives rise to additional challenges for existing federated learning schemes.
The aim of this project will be to explore how some of these challenges for existing Federated Learning schemes can be overcome in environments where the nodes (i.e. vehicles) are mobile.