The purpose of this paper is to present a calibrator to determine the best grid size of a Smart Surface. The Smart Surface is a micro‐electro mechanical systems (MEMS) whose goal is to sort micro‐parts.
The possible micro‐parts are rotated and translated, and their characteristics are stored in a database. Afterwards, when such a micro‐part is laid off the Smart Surface, its characteristics are compared to database values. Simulations show that some grid sizes are better than others in terms of success in part recognition.
The tests performed on all groups of three out of four models show that a sensors grid of (35, 35) is an appropriate parameter for the Smart Surface.
The authors plan to work on a more general case, using any kinds of parts.
The work allows the automation of the process of sorting micro‐parts, in assembly lines for example.
Few works exist for part recognition on very small parts and for choosing the best discretization scale.
