Skip to content

Development and Evaluation of a Quantitative Analysis System for Robot-assisted Musical Therapy


Jia-Yeu Lin, Yi-Hsiang Ma, Yean Han, Sarah Cosentino, Atsuo Takanishi


Autistic Spectrum Disorder (ASD) is a developmental disorder which impedes the normal development of social and behavioral skills and causes the affected subject to be anxious and unable to interact and communicate with others. Among the various forms of therapeutic approaches, music therapy is a commonly used method for improving their social skills. In our work, we developed a quantitative analysis system for accurate and timely feedback from the subjects, and aimed to create a high adaptability music therapy session with the use of the Waseda Anthropomorphic Saxophonist Robot No.5 (WAS-5). In the session, subjects are asked to perform a repetitive gesture to match the tempo of music played by WAS-5. Through an image processing algorithm for getting the human skeleton position, the tempo of the rhythmic movement can be calculated. WAS-5 will update the tempo accordingly to the feedback so that the subject can match their movement to the music tempo more easily. With the system, we expect to apply the robotic and motion capture technology in the field of psychology and child development which will allow for an adaptive and individually tailored therapy program with quantised evaluation standard.

Jia-Yeu Lin, Yi-Hsiang Ma and Yean Han are students at the Department of Integrative Bioscience and Biomedical Engineering, Waseda University, Japan. For correspondence: Sarah Cosentino is an Associate Professor at the Deptartment of Modern Mechanical Engineering, Waseda University, Japan. For correspondence: Atsuo Takanishi is a Professor at the Department of Modern Mechanical Engineering and a core member of Human Robotics Institute (HRI), Waseda University, Japan. For correspondence: The authors would like to express their gratitude to the Directorate General for Cultural Promotion and Cooperation at the Italian Ministry of Foreign Affairs for its support to RoboCasa. The authors would also like to express their gratitude to Tokyo Women’s Medical University, Waseda University Joint Institution for Advanced Biomedical Sciences (TWIns), Waseda University Program for Leading Graduate Schools, STMicroelectronics, Life Performance Research, Humanoid Robotics Institute (HRI), and SolidWorks for their support to this work.


Lx-Number Search

(e.g. A | 000123 | 01)

Export Citation