A projection-based approach for clarifying interaction partners in human-robot communication

Area of Science:

• Human-Robot Interaction
• Robotics
• Computer Vision

Background:

• Human-robot interaction (HRI) is crucial as robots integrate into daily life.
• Existing HRI methods often focus on humanoid robots and human-like gestures.
• Adapting interaction cues for non-humanoid robots presents unique challenges in conveying direct address.

Purpose of the Study:

• To propose a novel projection-based method for non-humanoid robots to initiate and enhance human interaction.
• To investigate if projected light can effectively signal direct address and improve user engagement.
• To assess the method's impact on perceived interaction clarity and participation in shared spaces.

Main Methods:

• Developed a projection system mounted on a non-humanoid robot.
• The projector illuminates the space around the robot and its interaction partners.
• Conducted experiments in two distinct scenarios to evaluate the method's effectiveness.

Main Results:

• The projection method successfully conveyed a strong sense of being directly addressed by the robot.
• Participants reported enhanced interaction clarity and a feeling of easier participation.
• The method proved effective even for individuals not actively engaged in the interaction.

Conclusions:

• Projection-based illumination is a viable and effective strategy for non-humanoid robots to improve HRI.
• This technique enhances the perception of direct address and encourages broader participation.
• The findings offer a new approach for designing more intuitive and engaging non-humanoid robot interactions.