Historical research guide · 0074 · 2010
DIRD Cognitive Limits on Simultaneous Control of Multiple Unmanned Spacecraft
DIA / AAWSAP contractor
In one minute
This is one of the thirty-eight Defense Intelligence Reference Documents Hal Puthoff administered for the Defense Intelligence Agency, and it sits at the crew-station end of the stack rather than the physics end. Writing in 2010, the author asks an operational question: if a deep-space mission flies as a fleet — a trailing craft of nuclear-powered electromagnets shielding the crew from radiation, halo craft carrying radar, mining and exploration craft, most of them largely automated — how many can one human pilot hold in mind at once? Because nothing had been published on remotely piloting several spacecraft, the report reviews the two closest bodies of evidence: forty years of air-traffic-control workload research, and the newer literature on flying several drones from one console. The number it settles on is 16 craft for simple destination selection, 7 for moderately complex piloting, and 4 when the craft are complex and unlike each other. It also shows that heart rate, skin conductance, blood flow and EEG can flag an overloaded operator in real time.
Connection
Chapter 1's point about the DIRD list is that the government commissioned an engineering reading list rather than a biology one, and this is the page of that list which simply assumes the fleet is already flying and asks who will crew it. It is the human-factors floor under the unified picture of chapter 13: propellantless craft still need someone holding the big picture.
From the source
The cognitive limit on the number of craft one pilot can simultaneously control is 16 for simple destination selection, 7 for moderately complex piloting or mission-task completion, and 4 for complex heterogeneous craft.Summary, p. iv; Chapter 6: Conclusions, p. 23
No current evidence shows that a complete mental picture can be maintained for more than about 16 objects at one time, even with external working-memory augmentation such as handwritten strip blocks or a second display.Chapter 6: Conclusions, p. 23; Chapter 5: Discussion, p. 22
Physiological variables can be employed objectively to indicate overload in real time: in a reanalysis of the Brookings air-traffic-control data, Wilson classified the overload condition consistently in more than 98 percent of cases.Chapter 3, p. 13
Automation that is right more than about 80 percent of the time degrades performance, because the operator stops confirming its alarms; a 20 to 25 percent false-alarm rate is optimal when the pilot is intended to work alongside the automation rather than rely on it.Chapter 3 (Wickens), p. 17; Chapter 5: Discussion, p. 22
There were zero peer-reviewed articles in the major journals on the remote piloting of multiple spacecraft published in the thirty years before 2010, so the whole question had to be approached through air traffic control and unmanned-vehicle analogues.Chapter 1: Introduction, p. 1
Within five years of spacecraft-specific simulator studies being commenced, the report expects those studies to appear in major peer-reviewed journals and expects physiological classification of operator overload to reach near-perfect accuracy.Chapter 5: Discussion, p. 22; Chapter 6: Conclusions, p. 23
The way in
https://documents2.theblackvault.com/documents/dia/AAWSAP-DIRDs/DIRD_34-DIRD_Cognitive_Limits_on_Simultaneous_Control_of_Multiple_Unmanned_Spacecraft.pdfHow to cite it
DIA / AAWSAP contractor (2010) DIRD Cognitive Limits on Simultaneous Control of Multiple Unmanned Spacecraft https://documents2.theblackvault.com/documents/dia/AAWSAP-DIRDs/DIRD_34-DIRD_Cognitive_Limits_on_Simultaneous_Control_of_Multiple_Unmanned_Spacecraft.pdf