Search

Models and modeling environments for human performance are becoming significant contributors to early system design and analysis procedures. Issues of levels of automation, physical environment, informational environment, and manning requirements are being addressed by such man/machine analysis systems. The research reported here investigates the close interaction between models of human cognition and models that described procedural performance. We describe a methodology for the decomposition of aircrew procedures that supports interaction with models of cognition on the basis of procedures observed; that serves to identify cockpit/avionics information sources and crew information requirements; and that provides the structure to support methods for function allocation among crew and aiding systems. Our approach is to develop an object-oriented, modular, executable software representation of the aircrew, the aircraft, and the procedures necessary to satisfy flight-phase goals. We then encode in a time-based language, taxonomies of the conceptual, relational, and procedural constraints among the cockpit avionics and control system and the aircrew. We have designed and implemented a goals/procedures hierarchic representation sufficient to describe procedural flow in the cockpit. We then execute the procedural representation in simulation software and calculate the values of the flight instruments, aircraft state variables and crew resources using the constraints available from the relationship taxonomies. The system provides a flexible, extensible, manipulative and executable representation of aircrew and procedures that is generally applicable to crew/procedure task-analysis. The representation supports developed methods of intent inference, and is extensible to include issues of information requirements and functional allocation. We are attempting to link the procedural representation to models of cognitive functions to establish several intent inference methods including procedural backtracking with concurrent search, temporal reasoning, and constraint checking for partial ordering of procedures. Finally, the representation is being linked to models of human decision making processes that include heuristic, propositional and prescriptive judgement models that are sensitive to the procedural content in which the valuative functions are being performed.

Some of the conceptual issues associated with pilot aiding systems are discussed and an implementation of one component of such an aiding system is described. It is essential that the format and content of the information the aiding system presents to the crew be compatible with the crew's mental models of the task. It is proposed that in order to cooperate effectively, both the aiding system and the flight crew should have consistent information processing models, especially at the point of interface. A general information processing strategy, developed by Rasmussen, was selected to serve as the bridge between the human and aiding system's information processes. The development and implementation of a model-based situation assessment and response generation system for commercial transport aircraft are described. The current implementation is a prototype which concentrates on engine and control surface failure situations and consequent flight emergencies. The aiding system, termed Recovery Recommendation System (RECORS), uses a causal model of the relevant subset of the flight domain to simulate the effects of these failures and to generate appropriate responses, given the current aircraft state and the constraints of the current flight phase. Since detailed information about the aircraft state may not always be available, the model represents the domain at varying levels of abstraction and uses the less detailed abstraction levels to make inferences when exact information is not available. The structure of this model is described in detail.

Telerobot systems are being developed to support a number of space mission applications. In low earth orbit, telerobots and teleoperated manipulators will be used in shuttle operations and space station construction/maintenance. Free flying telerobotic service vehicles will be used at low and geosynchronous orbital operations. Rovers and autonomous vehicles will be equipped with telerobotic devices in planetary exploration. In all of these systems, human operators will interact with the robot system at varied levels during the scheduled operations. The human operators may be in either orbital or ground-based control systems. To assure integrated system development and maximum utility across these systems, designers must be sensitive to the constraints and capabilities that the human brings to system operation and must be assisted in applying these human factors to system development. The simulation and analysis system is intended to serve the needs of system analysis/designers as an integrated workstation in support of telerobotic design.

Efforts to control remote manipulators in simulated microgravity described in report. Experiments conducted to determine effects of weightlessness on performance of operator controlling remote manipulator, or slave arm, by master arm at control station. Report concludes microgravity disturbs neuromotor control of human arm. Also suggests disturbance compensated for by adjustments in controller.

Experiments were performed at the Johnson Space Center (JSC), Manipulator Development Facility using the full scale Shuttle Remote Manipulator System (SRMS) to evaluate the effect of visual presentation through perspective display of the orthogonal forces and torques sensed at the manipulator end effector. The experiments investigated the effect of the display information on the management of forces and torques generated during payload berthing and deployment, as well as simulated satellite module change-out operations. The evaluation also addressed (1) issues of display format, including: force/torque scaling, point of resolution, and display mixing with video generated imagery, and (20) task related variables of payload size, alternative sources of guidance information, and control mode. The results are presented of a first-pass informal anlaysis of the analog, strip chart-recorded data from these evaluation tests. The results provide a relative measure of improvement in force management through the use of such a display, as well as information regarding the impact of display variables and task demands on operator performance.

This paper describes recent developments at JPL aimed at enhancing the operator's awareness of the remote task he or she is controlling in space teleoperation. The first development is related to the use of force-reflecting or bilateral manual control in weightless condition. The second development is related to integrated graphics displays of force and torque information originating from a remote robot hand. The third development is related to a distributed computational system in teleoperation organized in two groups: one integrated with a remote robot arm and hand, and another integrated with the control station. This computational system carries out interactive automation functions.

The effect of weightlessness on the human operator's performance in force reflecting position control of remote manipulators was investigated. A gravity compensation system was developed to simulate the effect of weightlessness on the operator's arm. A universal force reflecting hand controller (FRHC) and task simulation software were employed. Two experiments were performed because of anticipated disturbances in neuromotor control specification on the human operator in an orbital control environment to investigate: (1) the effect of controller stiffness on the attainment of a learned terminal position in the three dimensional controller space, and (2) the effect of controller stiffness and damping on force tracking of the contour of a simulated three dimensional cube using the part simulation of weightless conditions. The results support the extension of neuromotor control models, which postulate a stiffness balance encoding of terminal position, to three dimensional motion of a multilink system, confirm the existence of a disturbance in human manual control performance under gravity compensated conditions, and suggest techniques for compensation of weightlessness induced performance decrement through appropriate specification of hand controller response characteristics. These techniques are based on the human control model.

Teleoperation can be defined as the use of robotic devices having mobility, manipulative and some sensing capabilities, and remotely controlled by a human operator. The purpose of this paper is to discuss and exemplify technology issues related to the use of robots as man-extension or teleoperator systems in space. The main thrust of the paper is focused at research and development in the area of sensing- and computer-based automation from the viewpoint of man-machine interface devices and techniques. The objective of this R and D effort is to render space teleoperation efficient and safe through the use of devices and techniques which will permit integrated and task-level ('intelligent') two-way control communication between human operator and teleoperator machine in earth orbit.

Remote manipulators (teleoperators) are defined as mechanical devices which extend human manipulative ability to operational environments which are either hostile to or remote from the human operator. Teleoperation involves the explicit and active inclusion of the human operator (HO) in the system control loop. An investigation of human/teleoperator control interactions in the orbital operational environment has been conducted, taking into account the employment of a model describing human neuromotor control as a linear damped harmonic oscillator. The application of this model for the specification of the end point position in teleoperator control is explored, and two experiments are conducted. The results indicate the potential utility of relatively simple models of neuromotor control processes in investigating the interaction of the human operator and controller in teleoperation.

The concept of remote presence in telemanipulation is presented. A conceptual design of a prototype teleoperator system incorporating remote presence is described. The design is presented in functional terms, sensor, display, and control subsystem. An intermediate environment, in which the human operator is made to feel present, is explicated. The intermediate environment differs from the task environment due to the quantity and type of information presented to an operator and due to scaling factors protecting the operator from the hazards of the task environment. Potential benefits of remote presence systems, both for manipulation and for the study of human cognition and preception are discussed.