Affect recognition for interactive companions: challenges and design in real world scenarios

Abstract

Affect sensitivity is an important requirement for artificial companions to be capable of engaging in social interaction with human users. This paper provides a general overview of some of the issues arising from the design of an affect recognition framework for artificial companions. Limitations and challenges are discussed with respect to other capabilities of companions and a real world scenario where an iCat robot plays chess with children is presented. In this scenario, affective states that a robot companion should be able to recognise are identified and the non-verbal behaviours that are affected by the occurrence of these states in the children are investigated. The experimental results aim to provide the foundation for the design of an affect recognition system for a game companion: in this interaction scenario children tend to look at the iCat and smile more when they experience a positive feeling and they are engaged with the iCat.

Comparing task-based and socially intelligent behaviour in a robot bartender

“…Most current social robots play the role of a companion, often in a long-term, one-on-one relationship with the user-e.g., the systems described by Breazeal [5], Castellano et al [6], and Dautenhahn [7]. In this context, the primary goal of the robot is to build a relationship with the user through social interaction: the robot is primarily an interactive partner, and any task-based behaviour is secondary to this overall goal.…”

Social behavior recognition using body posture and head pose for human-robot interaction

“…Thus, the work of other groups, which focused on recognition of social signals in similar interactions as in JAMES, is better comparable to our work: Scheutz et al [3] presented a framework for a robot that recognizes and generates affective behavior and evaluated parts of it on a mobile robot receptionist and waiter. Castellano et al [4] described a system to recognize affective states and attentiveness. They found that sensing the intention and affective state of the human participants is crucial for HRI.…”

Human Body Pose Interpretation and Classification for Social Human-Robot Interaction

“…An overview of proposed techniques can be found in [53]. Only recently, a handful of researchers have also focused on recognizing a person's state or emotions via human body language [54][55][56][57][58][59][60][61][62][63][64]. For this discussion, we utilize the following definition of body language: the expression of emotions through human static body poses and/or body dynamics [56].…”