A hybrid approach to user-oriented programming of collaborative robots

The research reported in this paper proposes a new approach to collaborative robots that aims at improving the simplicity and efficiency of the programming task for non-technical users. It is grounded on three standpoints: (i) an elementary and disciplined paradigm for robot programming, called the simple programming journey, (ii) a hybrid interaction mode where robot tasks can be programmed using a natural language chat and, if necessary, can be completed and finalized through a block-based interface, and (iii) a robust cognitive match between the mental models of the user and the programming interface. The proposed approach has been implemented and tested through the development of a prototype programming environment called CAPIRCI, which can be tailored to different application domains through the definition of objects, locations, and actions. CAPIRCI has been tested by real users with a COBOTTA robot by DENSO WAVE Ltd. Two experimental tests have been carried out in order to validate the novel approach proposed and to assess its impact on end-user programming. The results obtained show that a hybrid approach exploiting both natural language dialogue and block-based interaction can help make the programming task easy and efficient for non-technical users.