Shared control between pilots and autopilots: an illustration of a cyberphysical human system

Abstract

This article considers the problem of control when two distinct decision makers, a human operator and an advanced automation working together, face severe uncertainties and anomalies. We focus on shared control architectures (SCAs) that allow an advantageous combination of their abilities and provide a desired resilient performance. Humans and automation are likely to be interchangeable for routine tasks under normal conditions. However, under severe anomalies, the two entities provide complementary actions. It could be argued that human experts excel at cognitive tasks, such as anomaly recognition and estimation, while fast response with reduced latencies may be better accomplished by automation. This then suggests that architectures that combine their action must be explored. One of the major challenges with two decision makers in the loop is bumpy transfer when control responsibility switches between them. We propose the use of a common metric that enables a smooth, bumpless transition when severe anomalies occur. This common metric is termed capacity for maneuver (CfM), which is a concept rooted in human behavior and can be identified in control systems as the actuator’s proximity to its limits of saturation. Two different SCAs are presented, both of which use CfM, and describe how human experts and automation can participate in a shared control action and recover gracefully from anomalous situations. Both of the SCAs are validated using human-in-the-loop experiments. The first architecture consists of a traded control action, where the human expert assumes control from automation when the CfM drops below a certain threshold and ensures a bumpless transfer. The second architecture includes a supervisory control action, where the human expert determines the tradeoff between the CfM and the corresponding degradation in command following and transmits suitable parameters to the automation when an anomaly occurs. The experimental results show that, in the context of flight control, these SCAs result in a bumpless, resilient performance.