Interaction
Situation-specific Social Behaviors
People's expectations and preferences about robots are influenced by the social and cultural context of interaction. The design of robot capabilities therefore has to incorporate diverse social cues and interaction structures, and be adaptable to different contexts. We study adaptive emotional expression, speech and voice characteristics, and how they contribute to people’s interpretations of the robot’s social agency, personality, and trustworthiness.
Robot Roles for Everyday Interaction
To be accepted in everyday interaction and become social companions, robots need contextually appropriate social roles and activities — a social niche. We use design research as empirical studies to identify robot roles and activities and to develop fitting robot behaviors. We also investigate how people react to and incorporate robots in their daily lives. Haru is currently being used in an inter generational daycare in storytelling and gameplay activities to address the community's goal of inspiring interaction between children and older adults. We are also exploring how Haru can be used in university dorms to support student social and emotional wellbeing.
Supporting Child’s Development and Creative Thinking
We are studying and developing capabilities for robots to evaluate and support collaborative problem-solving. This research also explores socio-emotional factors like embodiment, multi-modality, social presence, empathy and trust to support holistic human development in dynamic environments. Currently we are focusing on triadic interaction based on the tower of Hanoi task. In the later stages we will include high level cognitive skills
Robot Persuasiveness and Credibility
How can a robot’s verbal and non-verbal behavior make it more persuasive and credible in the social role it is in? Taking human interaction as a starting point, we explore functional effects of speech characteristics. These include prosodic delivery (speech melody, duration), the effects of timing and responsivity during turn-taking, and robotic displays of context awareness, as well as voice characteristics and non-verbal vocal behaviors that can help make the robot more credible.
The Integration of Multimodal Cues: Speech and Non-Verbal Robot Behavior
Multimodal action and interaction poses considerable challenges concerning the timing between an actor’s behaviors, and especially the timing of speech plays a special role since people bring in expectations from conversational interaction. We are currently investigating how the different expressive features of a robot can be coordinated in order to convey a single coherent message in all modalities, particularly in response to multimodal cues of human interaction partners.
Collaborative Human-Robot Interaction in Teams
Humans do not always make rational choices in games and are influenced by social factors and preferences, the feeling of fairness and cooperation. We investigate how these social factors are reflected in human-robot interaction by varying the level of cooperation and fairness in a robot and performing data analytics on the outcomes of the games in a collaborative games approach.
