Inverse Kinematics Learning By Modular Architecture Neural Networks With Performance Prediction Networks

Inverse kinematics computation using an artificial neu-ral network that learns the inverse kinematics of a robot arm has been employed by many researchers. However, conventional learning methodologies do not p a y enough attention to the discontinuity of the inverse kinemat-ics system of typical robot arms with joint limits. The inverse kinematics system of the robot arms, includ-ing a human arm with a wrist joint, is a multivalued and discontinuous function. Since it is difficult for a well-known multi-layer neuml network to approximate such a function, a correct inverse kinematics model for the end-effector’s overall position and orientation can-not be obtained b y the conventional methods. In order to overcome the drawbacks of the inverse kinematics solver consisting of a single neural network, we pro-pose a novel modular neural network architecture for the inverse kinematics model learning.