Parallel robots pose accuracy compensation using artificial neural networks

Abstract

Parallel robots pose accuracy compensation approach using artificial neural networks has been developed. In this method, an artificial neural network is used with conventional inverse kinematics computation module in parallel. A backpropagation neural network is designed and implemented to learn parallel robot kinematics model error. The trained neural network can be used to performed online pose accuracy compensation in task. Simulation and experimental results for a parallel robot are presented to show the effectiveness of the compensation method based on neural networks.