Vehicle Propulsion Information Integration

Marcus Klein and Lars Eriksson

Today's vehicles are under strict legislative requirements regarding emissions and thus need proper control and diagnosis functionality to secure operation within the prescribed limits. This will in the future include optimization, planning, and autonomous or semi-autonomous functions, but the focus of the project will be on information integration using observers, and on diagnosis techniques based on it.  A key characteristics is the interplay between modeling and information fusion, using information from in-vehicle sensors,  databases, as well as communicated information from other vehicles and infrastructure.  The core of the project is in information integration. One particularly interesting area is systematic and efficient analysis of combustion engine data. These have a direct impact on the performance and selection of control strategy. The data is very informative but to get reliable results, it is necessary to also combine the data with information from experts. Recent results concern the development and evaluation of systematic methods that combine traditional measurements with a parametric description obtained from expert knowledge in a systematic way. These methods have been applied to the problem of analyzing combustion engine data.