A Synergy-Based Motor Control Framework for the Fast Feedback Control of Musculoskeletal Systems

摘要: This paper presents a computational framework for the fast feedback control of musculoskeletal systems using muscle synergies. The proposed motor has hierarchical structure. A controller at higher level hierarchy handles trajectory planning and error compensation in task space. space only deals with task-related kinematic variables, thus is computationally efficient. output force vector space, which fed to low-level be translated into activity commands. Muscle synergies are employed make this force-to-activation (F2A) mapping explicit relationship between forces allows estimation activations that result reference force. synergy-enabled F2A replaces computationally-heavy non-linear optimization process by decomposition problem solvable real-time. performance evaluated comparing F2A-estimated activities against measured EMG data. results show algorithm can estimate kinematics/dynamics information ~70% accuracy. An example predictive simulation also presented; arbitrary movements 3D arm model quickly near-optimally. It two orders-of-magnitude faster than optimal controller, 12% increase compared optimal.