A transition in the art of piloting

In 1962, the task of building the human interface for the Apollo guidance system fell to Jim Nevins, an IL engineer with a background in control systems. For Mercury, NASA had turned the job over to psychologist Robert Voas and to ''human factors'' experts at McDonnell, but Nevins began looking for control systems engineers with experience in ergonomics. He found one in Tom Sheridan, an MIT assistant professor with expertise in mechanical engineering and psychology who would come to define ideas in supervisory control and telerobotics. The group eventually grew to about thirty people. They began with traditional aircraft controls, giving the spacecraft hand controllers and throttles, then adding the gyroscopes and an artificial horizon (the eight ball), and inputs to control pitch, yaw, and roll. Each solution raised more questions: how would such machines be operated in a weightless environment? How by people wearing bulky pressure suits? Could an astronaut in a helmet put his eye to the sextant?

Figure 7.6a, b

MIT Instrumentation Laboratory cartoon showing the extremes of automation. Too much automation (above) leaves the astronauts bored, awaiting an abort, while too little (right) overwhelms them with work. (Draper Laboratories/MIT Museum.)

Figure 7.6a, b

MIT Instrumentation Laboratory cartoon showing the extremes of automation. Too much automation (above) leaves the astronauts bored, awaiting an abort, while too little (right) overwhelms them with work. (Draper Laboratories/MIT Museum.)

Nevins's group drew large pictures with columns labeled ''astronaut'' and ''computer'' and ''ground,'' writing flow diagrams that detailed the transactions of data between these entities. ''It basically described the decision that the astronaut had to make, the question that he would put to the computer, the answer that the computer was expected to give'' and so on, Nevins said.55 The designers were less interested in the mathematics than in the interactions between the astronauts and the machines. They detailed these tasks in verbal form, coming up with numerous ''transaction schemes'' for each type of operation (figure 7.7).56

They began looking at the information flows. What did the astronauts need to know at different points in the flight? What would display that data? Where would it go? How big should the displays be? How many numbers needed to be displayed at any

Figure 7.6a, b

(continued)

Figure 7.6a, b

(continued)

one time? Should there be a picture tube to display data? Eventually they settled on three lines of numeric display, because a picture tube would be too heavy and consume too much power, and because engineers were used to seeing vectors that had three components, three dimensions.

Nevins articulated his philosophy for astronaut-computer interaction based on their fit into the Apollo system overall. Aware of the broader implications of this approach, he called it ''a transition in the art of piloting a vehicle.'' By ''transition,'' he meant that astronauts would rely on interactions with the computer, and with controllers on the ground, to an unprecedented degree that would change flight forever. ''The flight management system [in Apollo] instead of being an onboard operation is actually a highly integrated system of airborne and ground-based equipment.'' Overall, the astronauts and the computer formed but one part of ''a finely structured multilevel monitoring and decision process.'' Ground controllers, Nevins predicted, will feel as though they are inside the vehicle, monitoring for slow degradations and trends while

Figure 7.7

Interface flow diagram of a mid-course navigation illustrating information exchange between pilot and guidance and navigation system. (Draper Laboratories graphic 25381. Reprinted in Hall, Journey to the Moon, 62.)

Figure 7.7

Interface flow diagram of a mid-course navigation illustrating information exchange between pilot and guidance and navigation system. (Draper Laboratories graphic 25381. Reprinted in Hall, Journey to the Moon, 62.)

the pilots remain alert for events that require quicker reaction. Flying to the moon would be a long way from piloting in the classical mode (although Apollo paralleled new developments in aviation where pilots shared control with controllers on the ground).57

For Nevins, the crew played a critical part by managing systems and occasionally aligning the gyros. Their tasks consisted of (1) monitoring and decision making in guidance and control;(2) sequencing and initializing automated systems for guidance; navigation, and propulsion;and (3) optical pattern-matching tasks associated with tracking and identifying the stars for aligning the inertial system.

Even so, ''as technology improves ... many of these tasks will be automated,'' Nevins believed. By contrast, the computer would (1) monitor sensor data; (2) determine thrust times and vectors, trajectory parameters, and lines of sight;(3) maintain attitude control;and (4) guide the vehicles during thrusting maneuvers. Punching a few but tons would bring the spacecraft to a particular attitude; punching a few more would fire the main engine to propel it in or out of orbit. The computer's job looked more like traditional flying.

Because of the complexity of the task, Nevins continued, it would rely heavily on human control to adjudicate among flexible, redundant systems. Most critically, the crew would monitor the primary and backup systems and watch for indications of failure on one or the other. Nevins illustrated his philosophy with a detailed exploration of crew tasks during the final phases of landing, which we will examine in the following chapters. His approach amplified Chilton's original vision of astronauts as redundant backups, and placed a heavy burden on the crews in training and workload, especially when something went wrong. Nevins noted that the Apollo spacecraft had 448 switches and indicators, as opposed to 150 on Gemini and 102 on Mercury. Much of the crew's efforts would resemble operating a telephone switchboard as they flipped switches and actuated valves to configure the spacecraft for various phases of flight.

Ironically, Nevins seemed almost apologetic that the astronauts had to be so busy, explaining that only ''technological gaps'' required their involvement at all. Ultimately improvements in computers would automate the monitoring tasks and allow the human role to become ''more purely administrative, or supervisory.'' Future machines would ''relieve man of the necessity to play such an extensive role in either piloting or supervising his vehicle'' and would allow humans to spend more time doing scientific experiments or exploration.58 Of course, if the crews were no longer involved in the flying, NASA might send different kinds of people, perhaps scientists instead of test pilots.

Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook


Post a comment