Intelligent visual system of plastering robot

Abstract

The invention belongs to the technical field of electronic information, and specifically discloses an intelligent visual system of a plastering robot. The intelligent visual system of the plastering robot comprises a visual module 1, a visual module 2 and a controller connected with the visual module 1 and the visual module 2, wherein the visual module 1 and the visual module 2 are two same cameras; the controller is provided with a visual processing module used for processing the visual module 1 and the visual module 2, the visual processing module comprises an image segmentation unit and a posture estimation unit, the controller is further internally provided with a deep learning module, the deep learning module comprises a convolution layer, and the visual module 1 and the visual module 2 are further provided with laser ranging modules; and the visual modules and the visual processing module are used for photographing the robot and a surrounding environment thereof and calculating a specific position, and data are provided for the movement of the robot. The invention provides an intelligent visual system with an intelligent warning function, of a plastering robot.

Description

technical field [0001] The invention belongs to the technical field of electronic information, and in particular relates to an intelligent vision system of a plastering robot. Background technique [0002] Most traditional vision systems are based on monocular vision. At present, the monocular vision target tracking algorithm is relatively mature, and has achieved good development in many fields and aspects. However, monocular vision also has great defects. The amount of information in monocular vision is small, the three-dimensional information of the actual scene is lost during the projection process of the image, and the three-dimensional structure information of the tracking target cannot be fully utilized, so it has irreparable defects. When using methods based on monocular vision to track moving targets, there are often problems such as target occlusion and the interference of light changes and shadows in the surrounding scene. With the development and progress of th...

AI Technical Summary

Problems solved by technology

These robots are usually equipped with intelligent vision systems that allow them not only to detect parts in the workspace, but also to estimate their poses before taking further actions such as grasping, rotating, moving, assembling, etc. Object recognition and pose estimation It also plays an irreplaceable role in other important topics in the field of computer vision, such as virtual reality, augmented reality, autonomous driving, etc. With the vigorous development of deep learning technology and its good performance in visual tasks, convolution Neural networks have made major breakthroughs in all directions in the field of computer vision. The existing visual systems of plastering robots have the problem that the positional relationship is not fine enough and cannot be used for deep learning.

Images