Submitted Abstract
Mobile robots are expected to provide all kind of services for humans in various application scenarios and a dramatic increase of such service robot solutions is foreseen for the near future. However, in many of those scenarios the robots must be able to interact socially with people such as to respond appropriately to human behaviours and language, to learn and to collaborate with humans on human terms, as well as to act safely in the vicinity of humans. “Social robotics” aims to achieve this through development of social and communicative skills for physical robots and has become a very active own research area in recent years. While many research results exist in single specific areas that contribute to social robotics and while novel mobile robotic platforms offer considerable functionalities for the realisation of social robots at a comparatively low price, the efficient programming of social robots for a target application is still a very challenging problem. Most often the interdisciplinary integration of the different functionalities such as speech processing, gesture detection, computer vision etc. is solved in an ad-hoc manner for very specific problems, where knowledge and assumptions about the robot’s software remains implicit. In addition, human users show a wide range of possible behaviours creating a high level of uncertainty in the interaction. Furthermore, social robots often have to be developed together with domain experts for specific scenarios which are not robotic experts. One promising way for the programming of mobile robots in general is model-driven software engineering. Model-driven approaches are among the most prominent research topics in software engineering and hence several attempts of domain-specific modelling and languages are recently also proposed in robotics. However, these approaches are still in its infancy and also do not support the mentioned special aspects of social robots. Therefore, the main focus of this project is on the development of a model-driven software engineering approach for social robots and the proof of concept using a social robotic use case together with the City of Luxembourg as a project partner.