Using an RBF network with regularly-spaced position centres to approximate the inverse kinematic of a robot-vision system

This paper presents a novel solution using Radial basis function networks (RBFNs) to approximate the inverse kinematics of a robot-vision system. This approach has two fundamental principles: centres of hidden-layer units are regularly distributed in the workspace and constrained training data is used where inputs are collected around the centre positions in the workspace. To verify the performance of the proposed approach, a practical experiment has been performed using a Mitsubishi PA10-6CE manipulator observed by a webcam. All application programmes, such as robot servo control, neural network, and image processing tool, were written in C/C++ and run in a real robotic system. The experimental results prove that the proposed approach is effective.