Current students


Section: Systems and Control

Major Research topic:
Enhanced collaborative robotics via skill-based programming: the manipulation of linear deformable objects

The fourth industrial revolution, or Industry 4.0, led to a redefinition of industrial production in smart factories. In particular, the introduction of collaborative robots (cobots) brought significant changes in the paradigm of industrial robotics, mainly because of their flexibility that allows them to perform a wide range of tasks, at a moderate installation cost.

Although the implementation of this technology in assembly tasks might be relatively straightforward, different important topics must be addressed to fully exploit its potentialities. 

First of all, cobots require a collaborative framework to be effective, where humans and robots work together. It is, then, worthwhile to formalize and develop methodologies to optimally divide the operations of a complex task into two sets: the ones that humans can perform and those that robots must execute. 

In principle, the latter could be formalised with many different approaches. Among them, skill-based programming paradigms are the most promising and versatile, since they allow cobots to be easily and quickly reprogrammed, even by users with limited expertise in the field. This is due to the general descriptions of the operations, that are translated into competencies (the skills) that a given robot has and that can be, then, straightforwardly concatenated to describe and automatize complex tasks. 

Changing the programming approach is a first, yet crucial, step towards a full exploitation of the potentialities of collaborative robots. The real challenge, however, consists in building a rich set of skills: the larger the set, the more operations the cobots can perform and, hence, the higher the flexibility in the allocation of operations between them and human operators. 

This research will focus on one of the most challenging tasks in robotics: the manipulation of flexible linear objects, as cables. Its ubiquity in industrial applications makes it a fundamental skill that, however, requires a high level of dexterity, dealing with an under-constrained system. Deformable objects are, indeed, usually described by models with an infinite number of degrees of freedom, that introduce non-negligible uncertainties during the manipulation. 

All these issues will be addressed by this research from a methodological viewpoint, with systematic validation on industrially relevant robotic systems.