Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
DISCUSSION: CCYUENT5 ON TEIEERE5ENCE AND SUPERVISORY CONTROL Antal K. Bejc~y Telepresence and supervisory control technologies, as Professor Sheridan pointed it out, ~ ~ ~ ~ ~ existing teleoperator technology capabilities. . · . represent advancements or refinements of Both technologies are strongly driven By and rely upon Increased ccmpu ~ r and computing capabilities and are regarded ~= substantial contributors to evolving space station capabilities On the sense of reducing EVA astronaut involvement in assembly, servicing and maintenance operations. Mbreover, both technologies carry the premise of substantial spin-off for advancing capabilities of the U.S. production and service industries. Professor Sheridan and Professor Stark enumerated and elucidated many specific topics and issues in sensing, controls and displays for telepr-=ence and supervisory control which need research attention to advance the state of the art in the two technologies. In my discussion and comments, I would like to focus attention on the same research topics and issues from the following viewpoints: (a) In what sense and to what extent can we expect the enhancement of human operator capabilities through telepresence and supervisory control? (b) What specific conditions and constraints are impose] by the space application environment on the evolving telepresence and s ~ isory control technologies? (c) - ~ ~ ~ ~ ~ Ine muleldisclpllnary nature of the required research effort since neither telepresence nor supervisory control are ~ntrinsi~ly separate science or engineering disciplines. A brief description of the basic objectives of telepresence and supervisory control technologies may help illuminate the questions that arise from the above three viewpoints. The basic objective of telepr-=ence technology is to alleviate the human operator's sense of remoteness in the control station by providing sufficient information to the operator on the activities-= of ~~ ' ~ ' ~~ ~ the content of the last attribute He form" heavily dopers on human capabilities uTxi~r given conditions, on the capabilities arKicharacteristic~ of machines to tee controlled, and on the rapture of tasks to be ac~lished. Also implied In this technology is ache operator's enhanc ~ control response ability to the perceived remote events. Briefly, telepresence technology is aimed at providing--so to speak--a more intimate, the remote machine in usable form. 320
321 sensitive and high fidelity input and output connection between operator and remote machine. The basic objective of supervisory control technology is to provide sufficient capabilities for the human operator to tell the remote machine what to do and, eventually, how to do it, without involving the operator in continuous control coordination of a mwitibude of machine actuators needed to execute a task (note that a dual-arm system contains fourteen or more actuators). Thus, in supervisory mode of control, the operator con ~ ols the tack instead of controlling the individual degrees of freedom an] associated actuators of a multi-degree-of-freedom complex machine. Implied in this technology are two important technical capabilities: (a) flexible automation of actions of a multi-degree-of-fr=4~om complex mechanical system, and (b) flexible lar~uage-like or mer~u~type interface to, or interaction winch, the automated mechanical actions of a remote machine. Several notes should be added to the Objective descriptions of teiepresence and supervisory control technologies. _ First, none of then elongates the he an op ~ ator from the operation, but both change the Operator's function assignments and employ human capabilities in new ways. Second, both technologies promise the performance of more tasks with better results, but, in doing so, both technologies also make a close reference to human capabilities of operators who will use evolving new devices and techniques in the control station. Third, both telepresence and supervisory control technologies mate reference to evolving capabilities of other technologies like sensing, high performance computer graphics, new electro-mechanical devices, ccmputer-base] flexible automation, expert systems for Plann m a and error recovery' and so on. tied to rich multidisciplinary activities. _ , ~ ~ , _ _ _ ~ _ Thus, the progress in both technologies are _ ~ - Fourth, both technologies require the evaluation and validation of their results relative to the application environment. For space station scenarios, this implies the effect of zero-g an human operators, restricted local resources (like power, work volume, etc.) for a control station in Barth orbit, limited communication bandwidth and some communication time delay between a control station and remote machines, fragile and sensitive nature of space systems a t=leoperator machine will be work mg on, changes in visual conditions in Earth orbit relative to visual conditions on Earth, and so on. The above not==, together with the objective description of telepresence and supervisory control technologies, motivate a few important conclusions. First, the high fidelity, human operator referenced, man-machine coupling hardly worked on In t~lepresence technology--suggests we revisit anthropomorphic machine technology. The pr ~ reason for the revisit is not a declaration of some intrinsic optimality of anthropomorphic machines, but a recognition of their potentially easy and natural interface to human operators to physically extend the rich human manipulative capabilities, emtcdied in the dexterity of the human hand, to remote place=. One may visualize a back~rivable glove-¢ype device on the operator's hand connected through bilateral control to a controllable mechanical replica of the human hand equipped with same
322 sensing capabilities. This vision may not seen Coo singe Men capabilities of component technologies needed for the develc~nment of this anthr~rFihic machine are car~sider~. , Second, the performance of noncompetitive, sir=~r or un~ tee epperator tas ~ In space may benefit f The devel ~ nt of shared manual and automatic computer control techniques whenever application scenarios permit their use. -these techniques intend to combine the best attributes of human operators and computer control under restricted conditions. Third, the operator is facing a very rich environment in the control station in terms of decision, command, control and information , , processing even with increased telepresence and supervisory control capabilities. Due to the nature and tine scale of activities ~ telemanipulation, the cgerator's mental stable and readiness can be compared to an airplane pilot's functional situation Curing take-off or landing. thus, proliferation of control and Information hardware the control station does not serve the best interest of the human operator. m e more computer technology is employed at the control an] information interface in the control station in a clever way, the better off is the human operator to make control decisions efficiently. Fourth, the RSD effort for advancing Presence and supervisory control technologies should be acoompanled by systematic work on developing a human factors data base and models for understanding and utilizing the results of these evolving technologies. It is apparent from the nature of the ~ evolving technologies that the Emits or limitations rest not so much with the technologies themselves but with the human capabilities to absorb and use these technologies. Fifth, final evaluation and validation of Presence and supervisory control technologies for space station naturally require experiments and manifests in space whenever human perception, decision, control and other activities are influenced by space conditions. Simulations a ~ useful research and development tools, and they can pave the way towards performance evaluation and validation. But a comprehensive simulation of true space conditions on Earth for developing a human factors data base and models in t-lepresence and supervisory control technologies does not seem feasible. Professor Stark make a strong case for N~SA-University research in this arena. me benefits of ROSAUniversity connections in human factors research in the field of t=1epresence and supervisory condom can indeed be manifest through past and present examples. Particularly appear are cases when graduate scents caky out the experiments part of hear thesis research at NP~;A-su~rted, unique laboratories like Per, I, JSC, and so on, or ken scents set same working time at NOVA laboratories as Imperative scents or as academic part-time employees working on topics relate to their university sties.
