Human factors for robot safety assessment

Abstract

This article presents the results of a study on including human factors in assessment of the level of danger involved in a robot operation. A dynamic risk assessment method, which considers all effective factors in a collision, is essential for planning and control of the human-safe robots. In this regard, the visibility of the robot in the human eyes can significantly effect the probability of collision. To this end, we propose a risk index which not only considers the physical factors effective in a collision, but also includes the direction of eye gaze and human body orientation. Also two sensory systems for measuring these values are introduced. To this effect, a non-invasive brain-machine-interface prototype system which allows the simple control of a switch is presented. This system creates a feedback signal that is a function of the position of the human eyes for the modulation of the risk assessment. Initial results suggest possible benefits of using this approach in human-safe systems.