The success of rising service robots will rely largely on their ability to persuade people to use their services. Simple scenarios in which a robot conveys information to a human could be enhanced given a deeper understanding of persuasion in the context of human robot interaction. These robots can further increase their utility with moving around and being responsive to people. Robot furniture is an upcoming area of social robotics where the furniture itself acts as the minimal social robot. These robots have already shown success in interacting via non-verbal behaviors, however, previous work has seldom considered the persuasive capability of their behavior except in [1] [2].

Robot furniture is a growing area of robotics research, as people easily anthropomorphize these simple robots and they fit in easily to many human environments. Could they also be of service in recruiting people to play chess? Prior work has found motion gestures to aid in persuasion, but this work has mostly occurred in in-lab studies and has not yet been applied to robot furniture. This paper assessed the efficacy of four motion strategies in persuading passerbyers to participate in a ChairBot Chess Tournament, which consisted of a table with a chessboard and two ChairBots - one for the white team, and another for the black team. The study occurred over a six-week period, seeking passersby to play chess in the atrium of our Computer Science building for an hour each Friday. Forward-Back motion was the most effective strategy in getting people to come to the table and play chess, while Spinning was the worst. Overall, people found the ChairBots to be friendly and somewhat dog-like. In-the-wild studies are challenging, but produce data that is highly likely to be replicable in future versions of the system. The results also support the potential of future robots to recruit participants to activities that they might already enjoy.