ABSTRACT
Neural prostheses and motor control studies may find in computational simulation studies helpful aids. This work, focusing on human postural dynamics and control, developed a model that includes: rigid body mechanics, muscular contraction and neural excitation dynamics, and also an associated geometric musculoskeletal model. A methodology for controller design was established, using the LQR approach, and the pseudo-inverse matrix for distribution of control torques among redundant musculotendon actuators, employing also an inverse model of muscular contraction dynamics. Numerical results show some effects on simulations for certain initial conditions, with LQR weight matrix variations. Muscular coordination pattern is addressed.
