Mixed-initiative control for remote characterization of hazardous environments

Abstract

Remote characterization of high radiation environments is a pressing application area where robots can provide benefits in terms of time, cost, safety and quality of data. However, the DOE roadmap for Robotics and Intelligent Machines states that 'usability' may well prove to be the most challenging and yet crucial component of robotic systems for remote characterization and handling of radioactive and hazardous materials. In 2001, the INEEL successfully deployed a teleoperated robotic system coupled with a Gamma Locating and Isotopic Identification Device (RGL&IID) to characterize an area that had been closed to human entry for many years. This paper examines the human-robot dynamic of this teleoperated task and the limitations inherent to the master-slave strategy employed. Next, the paper outlines an innovative, mixed-initiative command and control architecture developed to address these limitations. The resulting, mixed-initiative control architecture retains the human in the loop, but interleaves multiple levels of human intervention into the functioning of a robotic system that can, in turn, scale its own level of initiative to meet whatever level of input is handed down.

Sharing and Trading in a Human-Robot System

“…to provide perceptual guidance; and the robot will respond to the human command by scaling its autonomy to drive the mobile platform according in the direction of the gaze (Hoppenot and Colle 2002). In both cases, trading is normally used in conjunction with sharing to let human and robot assist each other via the exchange of control and task information when both have problem performing the assigned task (Sheridan 1992;Lee 1993;Kortenkamp et al 1997;Bourhis and Agostini 1998;Fong et al 2001b;Bruemmer 2003). The basic questions in sharing and trading are as follows (Sheridan 1992): In sharing -"Which tasks should be assigned to human and which to the robot?"…”

“…The aim is to achieve mutual compensation of both the human's and the robot's individual weakness (Sheridan 1992;Hirzinger 1993;Lee 1993;Bourhis and Agostini 1998;;Fong et al 2001b;Hoppenot and Colle 2002;Bruemmer 2003). For instance, sharing of control and sharing of autonomy has often been described in both the literature of telemanipulation (Sheridan 1992;Hirzinger 1993;Lee 1993) and teleoperation of mobile robot (Bourhis and Agostini 1998;Fong et al 2001b;Hoppenot and Colle 2002;Bruemmer 2003). In telemanipulation, an example of sharing is the manipulation of a task where the compliance control is done by the robot automatically while position control is achieved by human's manual control (Hirzinger 1993;Lee 1993).…”

“…To this end, innovative paradigms have been proposed over the years to redefine the roles of human and robot from the traditional master-slave relationship (Hancock 1992;Sheridan 1992), such as to model the human as cooperator (e.g. Lee 1993;Bourhis and Agostini 1998;Fong et al 2001b;Hoppenot and Colle 2002;Bruemmer 2003) rather than just as the master controller of the robot. On the other hand, the slave robot is modelled in such a way that it becomes an active assistant Open Access Database www.i-techonline.com (e.g.…”

“…Bourhis and Agostini 1998;Hoppenot and Colle 2002) or partner (e.g. Fong et al 2001b;Lee 1993;Bruemmer 2003) of the human, supporting perception and cooperative task execution. To design a cooperative HRS based on the above paradigms, a basic research issue is to consider how to achieve cooperation via appropriate degrees of sharing and trading between human and robot which constitutes the main focus of this paper.…”

Blending human and robot inputs for sliding scale autonomy

“…Research in this area has been called adjustable autonomy, sliding scale autonomy and mixed initiative. For examples of work in this area, see [3][4][5][6].…”