An image-based visual servo control system based on an eye-in-hand monocular camera for autonomous robotic grasping

Abstract

This paper presents an image-based visual servo control system incorporating with an eye-in-hand monocular camera without using the interaction matrix as traditional method. Instead, we propose the modified image-based visual servo control system for monocular from point error and area error. By using the method derived from the point error and area error of features and one value of the simple depth estimation, we can find camera velocity whereas a traditional method we need to know every depth values for each feature for determining the camera velocity. In addition, a physical eye-in-hand robot system is developed and implemented in order to validate the proposed algorithm. The experimental results obtained from the systems show significant improvement in control performances.