Visual Servo Control by Optimizing Hybrid Objective Function with Visibility and Path Constraints

Abstract

This paper presents a new stable hybrid visual servo controller to complete the task of robot arm positioning. Our method is generic hybrid method and superior to the state of the art. The objective function is designed to include the full but weighted 2D and 3D available information. The positioning task has been formulated as a minimization problem. Here, each of the 2D and 3D error functions is used to control the six degrees of freedom. The importance weights are computed to satisfy a set of constraints defined by the visual servoing process. Stability analysis is also presented for the proposed control law. The experimental evaluation is done in order to show the enhanced performance of the visual servoing process. Simulation results show that this method provides an efficient solution to the camera retreat and features visibility problems. Performance of the visual servoing system is evaluated by its ability to keep features visible in the image and the Cartesian trajectory within the robot workspace during the process.