Integrated vision robot clamping jaw and using method thereof

Abstract

The invention discloses an integrated vision robot clamping jaw and a using method thereof. The integrated vision robot clamping jaw comprises a robot body and a clamping jaw body located at a moving end of the robot body, wherein the clamping jaw body comprises a control box, a vision module located on the control box and clamping jaw fingers located at the two ends of the control box, the vision module is located at the center position of the control box, the control view field of the vision module is located at the symmetrical positions of the two clamping jaw fingers, the vision module is electrically connected with the control box, and the clamping jaw fingers are electrically connected with the control box. According to the robot clamping jaw, the locating efficiency is increased; whether the clamping jaw is opened or closed to pick up an object, the relative pose of the object and the clamping jaw can be detected, and the grabbing precision is improved; the installation is facilitated; the matching speed can be increased; and the view field is symmetrical, so that when the relative pose of the object and fingertips is calculated, the precision is not easily influenced by the distortion of a camera, and the locating error is reduced.

Description

Technical field [0001] The present invention belongs to the technical field of robotic jaws, and more particularly to an integrated visual robot jaw and a method of use. Background technique [0002] In the industrial world, the robot is used for sorting, and the robot is generally used when gripping, and the positioning of the item is generally used in the position of the item and the calculation of the position. The specific implementation method is to bind the camera coordinate system with the robot coordinate system. After the visual detection target position, the target position is automatically converted to the position of the robot coordinate system, thereby performing automatic positioning crawler. . The visual sensor is divided into two kinds. One is that the camera is fixed outside the robotic arm, called as Eye-to-hand, one is the visual installation at the end of the robot, followed by the end of the robot, called Eye- IN-hand. In the traditional application, the visu...

AI Technical Summary

Problems solved by technology

In traditional applications, the visual sensor can only take a picture once. After obtaining the target position, it is handed over to the manipulator for independent grasping. It cannot monitor the relative position and posture of the manipulator's gripper and the object in real time. It belongs to open-loop control and has a large error. There are often crawling failures

At the same time, because the traditional camera is installed on the side of the gripper, the field of view of the camera is easily blocked by the arm and the gripper of the manipulator, so it is impossible to take multiple shots of the target and correct the pose error in real time

In addition, the mode that the vision sensor is installed on one side of the gripper also increases the overall volume of the gripper, which is not conducive to grasping objects in a narrow space

Images