A Hybrid Collaborative Operation for Human-Robot Interaction Supported by Machine Learning

Abstract

Speed and separation monitoring (SSM) as well as power and force limiting (PFL) are two of the four permissible collaborative operations in human-robot interaction (HRI). Current standards and guidelines provide users and system integrators with a simple basis to calculate permissible separation distances between human workers and robots. However, problems occur in practical implementations, as the safety zones are oversized due to various simplifications and corresponding path velocities have to be significantly reduced. This leads, for example, to cycle time losses and wasted space within the respective HRI application. The present paper describes an approach to combine the SSM and PFL collaborative operations and to exploit the optimization potential for robot motion planning and the design of collaborative workstations. To localize the necessary human body regions, we integrate a method from the field of machine learning. Using an exemplary HRI scenario with the PILZ lightweight robot PRBT, we validate the approach presented here and discuss initial results.