Information-Driven Multirobot Behavior Adaptation to Emotional Intention in Human Robot Interaction

To adapt robots' behavior to human emotional intention, an information-driven multirobot behavior adaptation mechanism is proposed in human-robot interaction (HRI). In the mechanism, optimal policy of behavior is selected by information-driven fuzzy friend-Q learning (IDFFQ), and facial expression with identification information are used to understand human emotional intention. It aims to make robots be capable of understanding and adapting their behaviors to human emotional intention, in such a way that HRI runs smoothly. Simulation experiments are performed according to a scenario of drinking at a bar. Results show that the proposed IDFFQ reduces 51 learning steps compared to the fuzzy production rule-based friend-Q learning (FPRFQ), and computational time is about 1/4 of the time consumed in FPRFQ. In Addition, the accuracy of emotion recognition and emotional intention understanding are 80.36% and 85.71%, respectively. The preliminary application experiments are carried out to the developing emotional social robot system, and the basic experimental results are shown in the scenario of drinking at a bar with three emotional robots and 12 volunteers.