Cyber-physical-social system between a humanoid robot and a virtual human through a shared platform for adaptive agent ecology

Abstract

Two artificial agents U+0028 a humanoid robot and a virtual human U+0029 are enriched with various similar intelligence, autonomy, functionalities and interaction modalities. The agents are integrated in the form of a cyber-physical-social system U+0028 CPSS U+0029 through a shared communication platform to create a social ecology. In the ecology, the agents collaborate U+0028 assist each other U+0029 to perform a real-world task U+0028 search for a hidden object U+0028 for the benefits of humans. A robot-virtual human bilateral trust model is derived and a real-time trust measurement method is developed. The role of taking initiative in the collaboration is switched between the agents following a finite state machine model triggered by bilateral trust, which results in a mixedinitiative collaboration. A scheme is developed to evaluate the performance of the agents in the ecology through the CPSS. The results show that the robot and the virtual human perform satisfactorily in the collaboration through the CPSS. The results thus prove the effectiveness of the real-world ecology between artificial agents of heterogeneous realities through a shared platform based on trust-triggered mixed-initiatives. The results can help develop adaptive social ecology comprising intelligent agents of heterogeneous realities to assist humans in various tasks through collaboration between the agents in the form of a CPSS.