Human Factors of Teleoperation in Space

Abstract

The burgeoning variety of existing and projected telerobot applications in hazardous environments has heightened human factors interest and concern about enhancing the utility and reliability of telerobotic systems. Use of telerobots in space applications is receiving particularly close attention. Current plans suggest that a significant telerobotic workforce may be employed for assembly, servicing, and/or maintenance tasks on Space Station Freedom (SSF). Future ambitious space missions, such as the lunar colony or the mission to Mars, also may entail extensive application of telerobotics in a variety of contexts.

The purpose of this symposium is to profile the human factors which may be expected to critically influence human performance with telerobotic systems in space environments. In the broadest sense, this question boils down to how well the operator can control the telerobot, in the face of different task requirements and operational conditions. To reap the anticipated benefits of telerobotic systems in space, related primarily to increased safety, efficiency, and productivity of task performance accompanied by reduced costs, it is essential that the control requirements for teleoperation are compliant with control capabilities and limitations of the human operator. The term workstation telepresence has been introduced to describe such human-telerobot system compliance, which enables the human operator to effectively project his/her body image and behavioral skills to control of the telerobot itself.

Accumulated evidence from human factors and human performance research suggests that a broad spectrum of factors may be expected to influence human variability in task performance with telerobots. These are summarized in Figure 1. The factors in Figure 1 may be categorized into three general areas: (1) task-specific design factors (to the left in the figure), related to the distinctive spatial and temporal properties and organization of different teleoperational tasks; (2) interface design factors, related to specific design features of the human-computer-telerobot interface, to the tracking requirements of the system, and to the modalities, sources (instrumental, operational, reactive), and spatiotemporal perturbations of sensory feedback introduced by interface design features; and (3) behavioral performance factors (to the right in figure), related to individual differences in the movement and sensory feedback control, perceptual, and cognitive skills and proficiency of the teleoperator, as they affect task execution.

This symposium offers four papers which collectively provide conceptual and experimental insight into many of the factors specified in Figure 1, as they affect teleoperation. The report by Sheridan compares and contrasts human performance in virtual versus actual environments, in relation to creating telepresence using visual, force, and tactile feedback from the telerobot. Two reports are concerned with ambitious plans by NASA to develop a telerobot workforce for SSF. The report by Stuart and colleagues deals with how the design of the hand controller interface influences teleoperation in relation to space-based applications. The Thomason and Yorchak report addresses the influence of various design and operator factors, including microgravity, on teleoperation in space with the Flight Telerobotic Servicer. The fourth paper in the symposium, by Draper, Handel, and Hood, is concerned with the influence of system and task design factors on the rate and fidelity of teleoperation. These authors use a standardized Fitts' tapping task to document the critical influence that task-specific design factors have on task execution with telerobots.

Two recurring themes are apparent in all of these reports. The first is that despite extensive engineering research and development on the technology of teleoperation, task performance with telerobots is substantially degraded relative to hands-on performance for most applications. The second is that the nature and degree of variability and decremental change in task performance during teleoperation appears to be highly specific to design characteristics of the system used and the task required. Investigation and resolution of both of these problem areas represents fertile grist for the mill of human factors research. In particular, findings in these reports suggest that realization of current plans for expanded application of remote telerobotic systems in space may be largely dependent upon improvements in task and interface design features, to achieve significant gains in system efficiency and safety. The remainder of this summary focuses upon system design factors that most critically influence performance variability during teleoperation, with particular attention to design-induced effects of spatiotemporal perturbations in sensory feedback.