Roles of Human Operator and Machine in Decision Aid Strategies for Target Detection

摘要

A common weakness in the design of detection systems is to envision the automation performing with the same probabilities of detection and false alarm as those expected of the overall detection system. This accounts for much of the dissatisfaction occasioned by conventional auto-detection. The situation is reviewed in light of the simple, common-sense principle of matching responsibilities to abilities for every detector, human or machine, in a larger detection system. Three basic design strategies for target detection, all satisfying this balance principle, are outlined in this work: (1) an advisory strategy to fill operator or incidental deficiencies; (2) an advanced sensor strategy to reduce operator uncertainty, and (3) an ergonomic strategy to keep the operator in top form. Military detection systems apparently demonstrate in microcosm many of the human-machine dynamics involved on a much larger, macroscopic scale in command and control. Detection systems may therefore be an ideal test bed for proving decision-aid principles on a smaller, more tractable scale, for use later at more complicated levels of command and control. The principle of matching responsibilities to abilities may generalize in this way.