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

Many nursing care robots have been developed to assist patients with sit-to-stand (STS) movement. However, little research has focused on user's negative psychological changes during STS movement when assisted by a robot. STS movement accompanied with a negative psychological change is defined as a fault. The main purpose of this study was to propose a method of conveying faults to a nursing care robot through the vertical ground reaction force (vGRF). Experiments on STS movement were executed five times with ten healthy subjects under four conditions: two self-performed STSs with seat heights of 43 and 62 cm, and two robot-assisted STSs with a seat height of 43 cm and end-effector speeds of 2 and 5 s. Subjects answered a questionnaire on how they felt under the four experimental conditions. Time series data on the vGRF were measured with a Wii Balance Board (WBB). A classifier was designed according to the data on the STS smoothness in the frequency domain. The results showed that the proposed classifier had a high probability of discriminating fault classes from others. Furthermore, faults were found to result in larger standard deviations of the peak values of smoothness. The center of mass trajectories of the human body under the same experimental conditions were used to crosscheck the experimental results. Then, the angles and angular velocities of the trunk and ankle were utilized to discuss the synchrony of the body segments. Other works on more advanced signal analysis and superior fault classification approaches were also discussed. It was concluded that faults in the assistance of nursing care robots can be detected from the STS smoothness by measuring the vGRF.