Pilot in the Loop

During the late 1950s and early 1960s, the value of a human pilot and redundant systems in space vehicles was a matter of some controversy. There was, and continues to be, a great debate on the relative merits of piloted vehicles versus automated ones. Because it had many similarities to early spacecraft, the X-15 became the subject of several evaluations and studies to determine whether the general design approach taken concerning redundancy and control were appropriate.-1146!

The AFFTC conducted one of these evaluations during late 1961. The basic approach of the study was to perform a detailed flight-by-flight engineering analysis of each problem or failure that occurred for the 44 free flights, plus an additional 2 captive flights and 30 aborts, made through 1 November 1961. For each problem, researchers assessed the action taken by the pilot or redundant system with regard to its impact on mission success and vehicle recovery. The researchers then compared the results with those that would be obtained on a hypothetical unmanned and/or non-redundant X-15.147

The researchers strictly adhered to several important ground rules during the evaluation and documented every problem, whether it seemed significant or not. The researchers conservatively assessed the benefits of the pilot-in-the-loop and redundancy to avoid any glorification of either of these elements.-148! The researchers also attempted to minimize conjecture, especially in the case of the hypothetical unmanned X-15. For instance, the researchers did not credit a pilot with detection or corrective action that some other element would definitely have provided in his absence. Likewise, he was not marked down for detrimental effects that would have been the same without a pilot. The study used a similar assessment scheme for redundancy. Finally, for the hypothetical unmanned X-15, the study assumed no changes to systems or components other than removing all redundant systems and substituting relatively simple and reliable present-day systems in their place.149

The results were not surprising. Of the 44 free flights conducted up to that time, researchers considered 43 successful as flown.150! Computed as an airplane that carried a pilot but no redundant systems, only 27 would have been successful. The number fell to 24 with redundancy but without a pilot, and to only 23 with no redundancy or pilot. The study noted that 19 flights were completely trouble – free, so they would have been successful in any of the three configurations. Significantly, the evaluation showed that the majority of times the mission was not

successful, the aircraft would have been lost. In fact, in the case of a piloted but non-redundant X-15, the study showed that 14 aircraft would have been lost in 44 missions.-1151

Around the same time, The Boeing Company conducted a similar study that analyzed the first 60 flights of the Bomarc surface-to-air missile. This large unmanned missile was designed to be relatively non-redundant in order to keep its manufacturing costs low.-1152 The Boeing study compared the actual flight results with a theoretical piloted Bomarc that incorporated a level of redundancy mostly equivalent to the X-15. Thus, the Boeing study was roughly the inverse extrapolation of the AFFTC X-15 evaluation, and the results bore an amazing similarity. For the X-15, the total mission success rate had been approximately 98%, which compared well to the computed 97% rate for a piloted and redundant Bomarc. Conversely, for both the actual Bomarc and the theoretical unmanned, non-redundant X-15, the total mission success rate was an identical 43%. This lent credibility to the idea that, with the current state of the art, it was still important to include a pilot in the loop.-153