Control, estimation and abstraction in fusion architectures: lessons from human information processing

Human information fusion offers important principles for the design of machine-based information fusion. The most important of these are the separability of control, estimation and abstraction, without which competent real-time responsiveness cannot be assured. Another principle is to separate intention from embodiment, so that intentions can be flexibly embodied. Yet another principle, similar to that of human emotions, is the use of positive feedback events, and subsequent feedforward, to produce timely responses to dynamic situations. Still another principle leveraging the trajectory of human cognitive development is to control the scope of processing by layering data abstractions both to manage bandwidth and so that the processing results can be immediately apprehended by human users. Using these principles, machine-based information fusion can increase both its competence and also the users' perception of its competence, increasing the likelihood that the machine will be received and treated as a trusted partner.