Dynamic Projection of Human Motion for Safe and Efficient Human-Robot-Collaboration

Abstract

In the modern manufacturing process, novel technologies enable the collaboration between humans and robots, which increases productivity by keeping flexibility. However, these technologies also lead to new challenges, e.g. ensuring safety for the human being in fenceless robot applications. State-of-the-art methods include, for example, the sensor technologies detect collisions to slow the robot velocities to have a low impact or to stop the robot while entering the shared workspace. Hence, the new research about the exploitation of a better performance in Human-Robot Collaboration (HRC) while maintaining the safety of the human attracts the attention. To achieve this goal, one strategy is to add an additional speed margin if a "worst-case" movement of the human can be predicted. In the context of this work, an approach of the dynamic human motion projection is proposed for typical assembly tasks. The human upper body is simplified as a five-degree-of-freedom (5-DOF) rigid-body model. A control-oriented projection model is proposed and its parameters are estimated from the experiment data of the human capability. In combination with a human-state estimator and a collision estimator, the "worst-case" motion is projected during the HRC scenario. This method is tested and validated in online and offline cases. In the end, the estimated collision time is used for calculating the increased speed limit of the robot, which verifies the improvement of the efficiency of HRC.