Imitating Human Strategy for Social Robot in Real-Time Two-Player Games

Chuanxiong Zheng, Hui Wang, Lei Zhang, Jiangshan Hao, Randy Gomez, Keisuke Nakamura, Guangliang Li

Abstract: In this paper, we build upon the latest advances in the field of gaming AI and social robots by developing a social robot Haru to learn and imitate human strategy in a real-time two player game. Results of our two preliminary user studies show that with our proposed framework, Haru is able to learn and imitate human strategies from different human players, and adapt strategies accordingly when the human player changes the game strategy. Moreover, our user study shows the preference of human players to our Haru which can adapt and imitate human strategy during game playing over Haru with fixed strategies. In addition, our results show that the level of fixed strategy seems to have an effect on the acceptance of social robot in game playing and the distraction of human player from playing the game will decrease Haru’s performance.