Improving Challenge/Skill Ratio in a Multimodal Interface by Simultaneously Adapting Game Difficulty and Haptic...

Area of Science:

• Robotics
• Human-Computer Interaction
• Affective Computing

Background:

• Harmonizing robotic devices with human users requires robots to perceive psychosomatic impacts from user emotional states like frustration or boredom.
• Existing robotic multimodal interfaces lack adaptive capabilities to modulate user-perceived challenge and skill during cooperative tasks.

Purpose of the Study:

• To introduce a novel biocooperative control architecture for robotic multimodal interfaces.
• To enhance the perceived challenge/skill balance for users interacting with robots in cooperative scenarios.
• To optimize the integration of physiological signals for real-time feedback.

Main Methods:

• Open-loop experiments identified heart rate, skin conductance level, and skin conductance response frequency as optimal physiological signals for feedback.
• A biocooperative controller with two degrees of freedom was developed, modulating game difficulty and haptic assistance.
• The controller utilized performance and psychophysiological feedback to adjust challenge and skill perception.
• A new metric, FlowIndex, was proposed to quantify the challenge/skill relation.

Main Results:

• The proposed biocooperative control architecture significantly improved the challenge/skill relation, as indicated by a higher FlowIndex.
• The system demonstrated an enhanced balance between augmented performance and user satisfaction, leading to a higher level of valence.
• Results showed superior performance and user experience compared to previously published comparable work.

Conclusions:

• The developed biocooperative control architecture effectively harmonizes robotic devices with human users by adapting to emotional states.
• The system successfully modulates perceived challenge and skill, leading to more enjoyable and satisfactory human-robot interaction.
• The FlowIndex metric provides a valuable tool for quantifying and visualizing the challenge/skill dynamic in human-robot interaction.