Dynamical and Complex Behaviors in Human-Machine Co-Adaptive Systems

Progressive system redesigns are needed with respect to human machine interactions to increase system reliability and transparency by increasing human-system interactions and especially a human user's proactive participation, rather than by eliminating the human out of the loop. Such a socially-centered view on the human-machine system design regards a human and an automated agent as equivalent partners, and through their mixed-initiative interactions some novel relations of mutual dependency and reciprocity would emerge as well as flexible changes of role-taking are expected. We develop a new idea called co-adaptive design principle, which means that both a human user and a machine should be able to adapt to the other through experiencing the interactions occurring between them. We applied this idea to an artifacts design of a robot tele-operation systems.