Traceable machine vision systems for digital industrial applications

Machine vision systems (MVSs) are crucial to many high-value industries where Europe is globally competitive, and to the European objectives in terms of digital transformation and green deal. But for these systems to achieve their full potential, further work is needed. Proposers addressing this SRT should establish the traceability of existing and newly developed MVSs combined with other measuring devices, develop digital twins (DTs) of MVSs based on data and physical driven models, and implement robust matching and analysis algorithms for large amount of recorded raw data. Additionally, the applicability of the developed methods and tools should be demonstrated through case studies and scenarios covering multiple industrial applications.

The specific objectives are to establish the traceability of existing and newly developed industrial MVSs used in i) dimensional quality, ii) surface quality, iii) structural quality, and iv) operational quality; develop DTs of selected and newly developed MVSs through physical models and/or computational models applying AI driven methods, and to predict their responses in analysing systematic errors, as well as to obtain the optimal measurements strategy in the shortest cycle time; implement methods for quantifying the uncertainty of the developed DTs for MVSs; investigate and evaluate novel methods and algorithms for dense image matching of multiple recorded images, using softgauges; To facilitate the take up of the technology, good practice guides and measurement infrastructure.