Fault classification with discriminant analysis during sit-to-stand movement assisted by a nursing care robot

Significance Statement

Sit-to-stand movement is important to everyday life, which is used to change from a sitting to a standing position. However, for the elderly, their capacity to perform this elementary task may weaken due to deterioration of muscle strength, balance and joint range of movement. This elevates the risk of improper functioning and mobility for activities of daily living, institutionalization and death. A reduction in nursing care specialists has caused an overwhelming need for sit-to-stand assisting robots.

There are a number of studies focusing on these robots, but only a few have focused on the psychological impact on subjects. Having a better understanding of these psychological issues may be helpful in facilitating better integration of medical gadgets with the users. An assisting robot is designed to provide satisfaction and if its assistance fails to offer this satisfaction, this may be considered a fault. A robot lacks the ability to know if its service satisfies the user or not. Therefore, inferring human psychology would be necessary in understanding human actions which would be central for human-robot interactions.

Recent advances in algorithms as well as sensors have improved robot’s perception ability, but this is insufficient as robot’s reaction depends on its understanding of human actions, therefore raising the concern of how to correlate psychological changes to legible data. For this reason, Osaka University scientists in Japan focused on conveying a fault to a nursing care robot through the vertical ground reaction force implementing a reliable yet economical method of providing rapid feedback. They collected psychological results by administering questionnaires to subjects. Their work is now published in Mechanical Systems and Signal Processing.

The authors selected ten human subjects for the study. None of the subjects reported use of medication, back pain, history of neurological disease, or lower limb pathology that may affect their standing balance. Also, there was only a small variation in their BMI. The researchers asked the subjects to answer a questionnaire on the sit-to-stand movement aspects that they felt strenuous or demanding for robot-assisted as well as self-performed experiments. This was meant to be a medium between physical movements and psychological changes. Based on the feedback from the questionnaire, the authors were able to define a selected class as a fault if accompanied by a demanding feeling.

Center of mass and the center of pressure of the human body were helpful in analyzing the human motion. Center of pressure was regarded as the projection on the ground plane of the centroid of the vertical force distribution. This could be obtained from a force plate directly.

Counting on the psychological changes, all subjects on the self-performed sit-to-stand, felt that the 43cm high chair was quiet demanding (class 2). For the robot-assisted movement, it was realized that adjusting the assisting speed affected the psychological change. All subjects, however, reported that they were supported when the assisting speeds was 2s. When the speed was lowered to 5 s, sit-to-stand became demanding. Therefore, class 4 was defined as fault.

The aim of the present study was to come up with an approach of informing a nursing care robot of the patient’s psychological change based on determining vertical ground reaction forces. The outcomes of the study indicated that the developed classifier had the ability to discriminate some faults classes from others.

Fault classification with discriminant analysis during sit-to-stand movement assisted by a nursing care robot- Advances in Engineering

About the author

Yuko Ohno received the Ph.D. degree in Institute of Medical Engineering from University of Tokyo, Japan, in 1985.  She successively worked as a research fellow in National Institute of Statistical Mathematics, Research Institute of National Cancer Center, and Tokyo Metropolitan Institute for Neuroscience.  Since 1995, she has been a professor in Osaka University Graduate School of Medicine, Suita, Japan.

Her research interests include Mathematical Health Science, Operation Metrics in Medical and Nursing service, Statistical analysis on the related Problem of Cancer Patients, Operation’s Research in Medicine. Now she is a vice president of the Japanese Society for Wellbeing Science and Assistive Technology, and Japan Society for Early Stage of Dementia, a board member of various academic societies in Japan.

About the author

Hieyong Jeong received the Ph.D. degree in Mechanical Engineering from Osaka University, Japan, in 2009. From Apr. 2009 to Nov. 2013, he was a Senior Research Engineer (Full-time) at SAMSUNG HEAVY INDUSTRIES, CO., LTD, Daejeon, Republic of Korea. From Nov. 2013 to Mar. 2014, he was a Researcher (Full-time) at Chonnam National University, Gwangju, Republic of Korea. From April 2014 to present, he has been an Assistant Professor (Specially Appointed / Full-time) at the Department of Robotics & Design for Innovative Healthcare, Graduate School of Medicine, Osaka University, Suita, Japan.

His research interests include Robotic hand manipulation, Human touch sense, Healthcare System, Human posture, Aging effect, and Industrial manipulator.

He received the Certificate of ROBOMEC from the R&D Division of the Japan Society of Mechanical Engineers in 2005, and also the Award of SICE Chugoku Branch from the Society of Instrument and Control Engineers in 2005.

Reference

Tianyi Wang, Hieyong Jeong, Mikio Watanabe, Yoshinori Iwatani, Yuko Ohno. Fault classification with discriminant analysis during sit-to-stand movement assisted by a nursing care robot. Mechanical Systems and Signal Processing, Available online 6 March 2017

Go To Mechanical Systems and Signal Processing

Check Also

Advancing Fusion Energy: High-Field REBCO Superconducting Magnets in the SPARC TFMC Program - Advances in Engineering

Advancing Fusion Energy: High-Field REBCO Superconducting Magnets in the SPARC TFMC Program