Visual Servo Control of Omnidirectional Mobile Manipulator

The mobile manipulator with vision sensor is of great significance to the future factory. How to make the mobile platform part participate in the process of trajectory planning efficiently, and how to reduce the camera calibration error and the influence of external factors on the visual servo accuracy are some problems that need to be solved. Uncalibrated Vision Servo (UVS) based on image has great flexibility and adaptability, therefore, it will solve many problems when it is applied to omni-directional mobile manipulator. A complete constrained mobile manipulator model is established. In this paper, we design a visual servo controller based on the centralized control strategy, and estimate the mapping relationship from the image feature space to the manipulator joint and the moving space of the mobile platform by online identification of the composite Jacobian matrix. Finally, the simulation experiment of Moving plane servo control in space is carried out.