The manufacturing industry is increasingly looking for process automation solutions capable of increasing efficiency and lowering production costs. For what concerns the variety of applications and the possible benefits that may result from a reliable application in the manufacturing industry, one of the areas of particular focus is the picking of randomly arranged items from a bin carried out by anthropomorphic manipulators, commonly referred to as "Bin Picking".

The automation of machine tools loading currently depends on the use of a completely flexible tool, that is a robot, and an extremely rigid mechanical periphery. The robot can indeed position its extremity anywhere in the work volume with an arbitrary orientation, whereas the mechanical periphery, which is usually made up of vibrators equipped with trap paths is designed to position and orientate a single component. Such rigidity, as well as the need to shake the pieces in order to generate micro-movements and move them, show this approach is not the best solution.

The implementation of a system capable of depalletizing randomly oriented parts by means of an anthropomorphic manipulator would definitely overcome such problems, thus providing an extremely effective instrument in the search for the optimization of production processes.

Early examples of bin picking go back to the late '90s, demonstrators which effectively carried out BIN PICKING on simple components have been presented several times. In spite of such examples, after more than a decade, BIN PICKING still remains a key challenge for the field of vision applied to robotics. What is then the reason which prevented the large-scale deployment of a solution offering such great advantages? A further analysis of the issue, as well as experience, have indicated how numerous and manifold the problems which need to be resolved are:

-difficulties in the recognition of the part

-low flexibility in the acquisition of the new parts

-overlap between the parts which are to be picked up

-interference between the parts in the bin and the gripping organ

-difficulties in reaching the gripping point, due, for example, to the pose of the pieces.

This whole range of serious issues which need to be dealt with using an equally wide range of skills and competences, has seriously limited the diffusion of bin picking systems, thus discouraging many system integrators who did not possess such a wide range of knowledge and expertise from proposing this kind of solutions. Although at present there can be no universal solution to the problem, a careful analysis of every single issue at stake and the right choice of technological solutions can lead to success in an extremely high number of cases.

3D CPS was developed to provide robotic system integrators with an instrument capable of autonomously managing all the problems named above, thus allowing for the development of BIN PICKING applications in an autonomous, rapid and effective way.


Discovered the 3D CPS:

3D CPS 4.0

3D CPS.2.0