Feedback self-learning industrial robot and working method thereof

Abstract

The invention discloses a feedback self-learning industrial robot and a working method thereof. The feedback self-learning industrial robot comprises a robot body, a robot joint, a robot end executor, a pressure sensor, a base, a vision sensor, a support and a robot base, wherein the robot end executor is arranged at the tail end of the robot body; the pressure sensor is mounted on the robot end executor; the vision sensor is fixedly mounted at the top of the support; and relevant actions are performed after a neural network is constructed, images are collected and processed, and the robot is controlled. According to the feedback self-learning industrial robot and the working method thereof, whether the end execution is successful or not is judged through the data from the pressure sensor, which provides a learning outcome basis for the self-learning neural network and further provides a learning way for the self-learning of the industrial robot; besides, according to the image processing result, the hand-eye relation calibration between a camera and the robot is not needed during the overall process, the self-adaption of related tasks of unknown objects is facilitated, and the application range is greatly expanded.

Description

technical field [0001] The invention relates to the field of industrial robots, in particular to a feedback type self-learning industrial robot and a working method. Background technique [0002] Traditional industrial robots mainly control the movement of robots through pre-programmed methods to achieve repetitive work. Robots cannot judge and work without the associated tasks being defined and programmed in advance. On this basis, by calibrating the relationship between the robot's visual sensor (eye) and the robot's end effector (hand), that is, the hand-eye relationship, combined with image processing methods to realize automatic hand-eye robots that work automatically have been widely used and developed. . This kind of robot is suitable for work such as gripping and handling of known object forms, and cannot automatically learn and operate on unknown objects. Contents of the invention [0003] The object of the present invention is to provide a feedback type self-l...

AI Technical Summary

Problems solved by technology

This kind of robot is suitable for gripping and handling of known objects, and cannot automatically learn and operate on unknown objects

Images