Abstract
The aim of our study was to develop a method by which a social robot cangreet passersby and get their attention without causing them to sufferdiscomfort.A number of customer services have recently come to be provided bysocial robots rather than people, including, serving as receptionists, guides,and exhibitors. Robot exhibitors, for example, can explain products beingpromoted by the robot owners. However, a sudden greeting by a robot can startlepassersby and cause discomfort to passersby.Social robots should thus adapttheir mannerisms to the situation they face regarding passersby.We developed amethod for meeting this requirement on the basis of the results of relatedwork. Our proposed method, user-centered reinforcement learning, enables robotsto greet passersby and get their attention without causing them to sufferdiscomfort (p<0.01) .The results of an experiment in the field, an officeentrance, demonstrated that our method meets this requirement.