Protection from Human Error: Guarded Motion Methodologies for Mobile Robots

Abstract

Industrial manipulators and unmanned systems often address a large number of tasks with some type of human-in-the-loop method. In these systems, the robot is given responsibility for some portion of the control tasks, but the human has some role for a variety of reasons; for example, the current technology may not be sufficient for the robot to complete the entire task: there may be safety, liability, or regulatory constraints, or the economics favor a human-in-the-loop process. An example of where human-in-the-loop control is of increasing interest is for telecommuting by health-care providers [1] and the general public [2] and for data gathering for disaster response [3]. These remote presence applications allow humans to perceive and act from a distance through a mobile robot. Remote presence is more challenging than telesurgery and space telepresence from an interface perspective, as the operators are not expected to be highly trained on robots and will be working in dynamic or unpredictable environments.