Generation of Bystander Robot Actions Based on Analysis of Relative Probability of Human Actions

Abstract

This paper describes a method of rule extraction for generation of appropriate actions by a robot in a multiparty conversation based on the relative probability of human actions in a similar situation. The proposed method was applied to a dataset collected from multiparty interactions between two robots and one human subject who took on the role of supporting one robot. By computing the relative occurrence probabilities of human actions after the execution of the robots’ actions, twenty rules describing human behavior in such a role were identified. To evaluate the rules, the human role was filled by a new bystander robot and other subjects were asked to report their impressions of video clips in which the bystander robot acted or did not act in accordance with the rules. The reported impressions and a quantitative analysis of the rules suggest that the behavior of listening and the supporting role that the subjects play can be reproduced by a bystander robot acting in accordance with the rules identified by the proposed method.

Generation of Head Movements of a Robot Using Multimodal Features of Peer Participants in Group Discussion Conversation

“…For example, Leite et al [33] used two MyKeepon robots in a scripted interactive storytelling system with a group of children. Vazquez et al investigated the effects of the robot's orientation and gaze direction in a conversation where the participant group has a brainstorming discussion about how to solve the robot's problem [34], bartender robots serving multiple customers in a bar [35][36][37], and a receptionist robot [38]. In the authors' work, the aim is to make the robot join the discussion and play the same role as the other peer human (or robot) participants.…”

“…Although not for multiple users, there are also works on generating a robot's head motion with a data-driven method. Sakai et al [37] derived the relative probabilities of the robot's head motion from a data corpus where one human user talks to two robots. Sakai et al [42] developed a head motion model based on the recognized speech acts of the operator of a tele-operated robot.…”