Smart Robotics Prosthesis Using Deep Learning and Musculoskeletal Modeling

摘要

With the application of deep learning, prosthetic rehabilitation can be carried out in a manner that not only emulates human mechanical skills and performance, but can work more efficiently. In this study, we introduce computer vision capability for a rehabilitation robot using a convolutional neural network. The human motion of scooping is studied by dividing it into four motion primitives or sub tasks. For each primitive, optimum human posture is identified in terms of muscular effort. Human motion skills are analyzed in terms of physiological parameters, including wrist pronation-supination angle, elbow flexion angle, shoulder rotation/abduction/flexion angles, and hand acceleration by musculoskeletal modeling. This analysis identified how humans execute the same activity for eight different materials. Optimum human motion for each material is mapped to a robotic arm with six degrees of freedom, which is equipped with a camera. Consequently, the activity can be performed efficiently based on human intuition in a dynamic environment.