“…In this context, the robot must infer the user intention, to interact more naturally, from the human perspective [8,9,10]. To this end, both visual (e.g., based on Microsoft Kinect TM [11]) and force feedback, have been used [12,13,14,15,16]. Generally, we believe that direct sensor-based methods, such as visual servoing [17], provide better solutions, for intuitive HRI, than planning techniques, requiring a priori models of the environment and agents [18].…”