Situation-adaptive autonomy: the potential for improving takeoff safety

Abstract

This paper discusses how authority can be traded between human and automation flexibly and dynamically in a situation-adaptive manner. Aircraft's rejected takeoff problem is taken as an example where a more sophisticated human-automation collaboration may be required. Upon an engine failure during the takeoff run, a pilot must decide whether to continue the takeoff or to abort it. Just a few seconds are allowable for identifying the situation, making decision, and applying appropriate controls. It is shown that the situation-adaptive trading control of authority is indispensable for attaining system safety under high time-criticality.