A visual control method of industrial manipulator based on deep convolutional neural network

Abstract

The invention discloses a visual control method of an industrial manipulator based on a deep convolutional neural network, comprising the steps of: 1) collecting and preprocessing visual information of a target object; 2) training and adjusting a deep convolutional neural network model; 3) verifying the model and save the model. The invention combines the deep convolutional neural network to extract the ideal grasping position of target objects with different postures, which improves the applicable range of the system, thereby overcoming the problem of poor recognition of specific target objects by traditional visual control, and effectively simplifying the difficulty of using industrial manipulators , providing a new method for industrial manipulator control, with good scalability.

Description

technical field [0001] The invention relates to the field of industrial manipulators, in particular to a visual control method for industrial manipulators based on a deep convolutional neural network. Background technique [0002] In industrial production, industrial robotic arms can replace manpower for simple and repetitive tasks, such as: picking, assembling, welding, packaging, beating, cutting, grinding, dragging and other production operations. Especially in dangerous and harsh working environments, robot technology is used to reduce potential safety risks. Research on robot technology is an important way to realize intelligent manufacturing and reduce production costs of enterprises. [0003] "Motion planning" and "task determination" are two key technologies for industrial robotic arms. "Motion planning" can be divided into two parts: path planning and trajectory generation. The purpose of path planning is to find a series of non-interfering path points for the in...

AI Technical Summary

Problems solved by technology

But the problem is that these sensors are not very accurate and cannot be used in industrial production in the short term

[0009] 3. Voice teaching: using predefined voice commands to manipulate the movement of industrial robots, the problem is that the operability accuracy is poor, and fine work cannot be performed well

[0011] The disadvantages of the above four ways of using industrial robots are that industrial robots need to operate according to predefined programs, require professionals to maintain, and are highly complex when applied to new tasks.

Difficulties in deployment and implementation have largely limited the growth of the robotics industry

Images