The increasing complexity of automotive systems and the high number of variants make the exploitation of available degrees of freedom the longer the more difficult. Model based control is generally advocated as a solution or at least as a support in this direction, but modeling is not a simple issue, at least for some problems in automotive applications, e.g. in emissions based control. Classical first principle methods are
extremely helpful because they provide a good insight into the operation of the systems, but frequently require too much effort and/or do not achieve the required precisions and/or are not suitable for online use. There is much know how available in the identification community to improve this, either by purely parameter estimation based approaches or by mixed models, but this know-how is frequently not tailored for the needs of the automotive community, and the cooperation is actually quite limited.
Against this background, this workshop aims at bringing together a limited number of potential users, in particular from the industry, identification experts, in particular from the academy, and successful users from both worlds, to discuss the issue and in particular to look on how to bring the two communities nearer.