Interactive learning in human-robot collaboration

In this paper, we investigated interactive learning between human subjects and robot experimentally, and its essential characteristics are examined using the dynamical systems approach. Our research concentrated on the navigation system of a specially developed humanoid robot called Robovie and seven human subjects whose eyes were covered , making them dependent on the robot for directions. We compared the usual feed-forward neural network (FFNN) without recursive connections and the recurrent neural network (RNN). Although the performances obtained with both the RNN and the FFNN improved in the early stages of learning, as the subject changed the operation by learning on its own, all performances gradually became unstable and failed. Results of a questionnaire given to the subjects confirmed that the FFNN gives better mental impressions , especially from the aspect of operability. When the robot used a consolidation-learning algorithm using the rehearsal outputs of the RNN, the performance improved even when interactive learning continued for a long time. The questionnaire results then also confirmed that the subject’s mental impressions of the RNN improved significantly. The dynamical systems analysis of RNNs support these differences. 1. Introduction Many kinds of the mechanical systems that cooperate with human beings have recently been studied in efforts to improve task performance and the mental impression of the persons using the system. A humanoid robot, for example, will not only have to help people work but also have to establish a new relationship with people in daily life. We focused on interactive learning between a human operator and a robot system, in a fundamental form to design a natural human-robot collaboration. It consists of the robot system, which learns the task including a human operator, and the human, who learns the task including the robot system. However, it is usually difficult to stabilize the system for a long period of time of operation because the incremental learning of such coupled and nested systems between humans and robots tends to generate quite complex dynamics. Although there have already been some studies of learning systems in man-machine cooperation [1][2], most of them only focused on short period operations in which the coop