Shared control based on roles for telerobotic systems

The classic approach to teleoperation system design seeks to augment human capabilities and allow the execution fo tasks in a remote (possible hazardous) environment. Recent developments have provided robotic systems with additional autonomy, thus allowing the possibility of having collaboration between humans and robots in local scenarios. This paper presents preliminary ideas where the concept of human-robot collaboration is extended to the area of robotic systems operated remotely. Thanks to the studies made to ascertain how humans interact when performing physical tasks in a collaborative manner, it is possible to use such concepts to study human-robot interactions and create more trasnparent teleoperation systems. The idea of using role-based strategies to perform tasks via remote robotic interfaces is explored, where different interaction models are analysed via the observation of how the “role” of the interacting agents evolves as the task is being perfromed. Additionally, the paper presents some of the research areas that have provided the basis of the proposed approach, as well as the main challenges for the application of such teleoperation system. It is noteworthy to mention that one of the main challenges is to provide the system with adaptive and learning strategies that allow the agents to dynamically specialize and provide a more natural human-robot interaction.