摘要

Seeking other joint trajectories of human arm in terms of given trajectory of hand is a wide research problem. In this paper, the computational problem of inverse kinematics of arm movement was investigated. On the basis of an eight-DOF arm model, an inverse kinematics algorithm recreating arm motion was developed in accordance with the human's feedback control mechanism of motor. In this method, the comfort level, an ergonomic index was adopted as the performance function of the motor control. Through maximizing the comfort level, human arm movement can be simulated when the end-effecter motion is given. This method combines human dynamics and ergonomics. To evaluate the approach, some arm goal-directed movement experiments were carried out. The joint trajectories were predicted with an average R-2 = 0.9627 and mean residual error MRE = 0.0094 m, which indicates that the presented algorithm provides a feasible method to simulate arm goal-directed movement and can be used as an efficient postural manipulation tool for the virtual reality.