Intelligent Robots for Use in Hazardous DOE Environments

Abstract : In Robotics and Intelligent Machines in the U.S. Department of Energy: A Critical Technology Roadmap, the DOE has identified the potential for Robots and Intelligent Machines (RIM) to greatly reduce cost, improve worker health and safety, augment product quality and increase overall productivity. In its long-term vision, the DOE has predicted that such RIM capabilities will be as pervasive and indispensable for the DOE and national interests as the PC is today. However, for this vision to be realized, critical issues pertaining to the interaction of humans and intelligent machines must be further explored and new technologies developed. In terms of time, cost and safety, 'usability' may well prove to be the most crucial component of RIM systems for remote handling of radioactive and hazardous materials and a wide variety of other operations. In this paper we examine the metrics used by the DOE to compare baseline radiation survey techniques with a teleoperated robotic survey recently conducted at the Idaho National Engineering and Environmental Laboratory (INEEL). Further, the paper discusses the difficulties and limitations of teleoperation evident from this deployment. To meet the current and future goals of the DOE, it is absolutely necessary to move beyond teleloperation and develop robot intelligence that can be interleaved with human intelligence to mitigate these difficulties. In response to this need, the INEEL has developed a mixed-initiative robotic system which can shift modes of autonomy on the fly, relying on its own intrinsic intelligence to protect itself and the environment as it works with human(s) to accomplish critical tasks.