Enhancing Emotional Expressiveness in Biomechanics Robotic Head: A Novel Fuzzy Approach for Robotic Facial Skin’s Actuators

Nguyen Minh Trieu, Nguyen Truong Thinh^*
Institute of Intelligent and Interactive Technologies, University of Economics Ho Chi Minh City–UEH, Ho Chi Minh City, 72516, Vietnam
* Corresponding Author: Nguyen Truong Thinh. Email: thinhnt@ueh.edu.vn
(This article belongs to the Special Issue: Applied Artificial Intelligence: Advanced Solutions for Engineering Real-World Challenges)

Computer Modeling in Engineering & Sciences https://doi.org/10.32604/cmes.2025.061339

Received 22 November 2024; Accepted 31 January 2025; Published online 17 March 2025

Download PDF

Abstract

In robotics and human-robot interaction, a robot’s capacity to express and react correctly to human emotions is essential. A signiffcant aspect of the capability involves controlling the robotic facial skin actuators in a way that resonates with human emotions. This research focuses on human anthropometric theories to design and control robotic facial actuators, addressing the limitations of existing approaches in expressing emotions naturally and accurately. The facial landmarks are extracted to determine the anthropometric indicators for designing the robot head and is employed to the displacement of these points to calculate emotional values using Fuzzy C-Mean (FCM). The rotating angles of skin actuators are required to account for the smaller emotions, which enhance the robot’s ability to perform emotions in reality. In addition, this study contributes a novel approach based on facial anthropometric indicators to tailor emotional expressions to diverse human characteristics, ensuring more personalized and intuitive interactions.The results demonstrated howfuzzy logic can be employed to improve a robot’s ability to express emotions, which are digitized into fuzzy values. This is also the contribution of the research, which laid the groundwork for robots that can interact with humans more intuitively and empathetically. The performed experiments demonstrated that the suitability of proposed models to conduct tasks related to human emotions with the accuracy of emotional value determination and motor angles is 0.96 and 0.97, respectively.