Human in the loop online estimation of robotic speed limits for safe human robot collaboration

Abstract

Collaborative robots (CR) gained increasing attention since system costs including costs for safety measurements are smaller, they are easier to program and they are more flexible than conventional industrial robots. In order to make use of the increased flexibility human-robot collaboration (HRC) robots might be applied flexibly. Therefore, not only reprogramming but also risk assessment has to be performed repeatedly. New safety standards relevant for HRC (DIN ISO 10218-1&2, ISO TS 15066) increase the complexity of the procedure that does not only involve the hazard identification, but also time-consuming force and stress measurements. The results are highly dependent on the current safety-settings of the robot, its current joint states and the contact point position. In order to speed up this process and make it more transparent, the computer aided safety assessment (CASA) tool aims to support during risk assessment. To achieve this, a visualization and simulation framework is developed, that makes use of accurate dynamic models of the robot and the body regions to estimate possible contact forces. Moreover, the geometry of the collaborative workcell is considered to highlight application specific hazards. The human factor is taken into consideration by online digitization of the co-workers’ body parts positions (key points).