Target positioning method and device based on depth vision sensor and automatic grabbing robot

Abstract

The invention provides a target positioning method and device based on a depth vision sensor and an automatic grabbing robot, and relates to the technical field of machine vision. According to the target positioning method, complex background interference information is filtered out through a depth vision sensor and a depth segmentation method, a target object in a scene is positioned, and the spatial position of a target is obtained. The automatic grabbing robot carries out kinematics planning on the track of the mechanical arm according to the spatial position information of the target, a motion instruction is transmitted to a single-chip microcomputer controlling the mechanical arm, and finally autonomous grabbing of the target by the tail end mechanical claw is achieved. The target object is quickly recognized through depth segmentation, interference of irrelevant background information is reduced, and the positioning accuracy is improved. The sensing capacity of the mechanical armis improved, the motion trail is autonomously planned according to the target point position, the target object in the working range is autonomously grabbed, and the motion trail does not need to bereprogrammed when the position of the target object is changed.

Description

technical field [0001] The invention relates to target recognition and positioning and mechanical arm kinematics planning, in particular to a target recognition method based on a depth vision sensor and the structural design of an autonomous grasping robot. Background technique [0002] Grasping or manipulating objects is a task often performed by robotic arms. For the robot arm, the recognition and positioning of the target object is the prerequisite for its successful grasp of the target object. However, general intelligent robots cannot easily identify and locate the target object in a complex environment like humans, usually through image processing. Algorithms are used to extract object features for recognition, but it has always been a difficult problem to accurately and quickly recognize targets in complex environments. [0003] In addition, the traditional robotic arm usually grasps the target according to the established process, and cannot receive external informa...

AI Technical Summary

Problems solved by technology

For the robot arm, the recognition and positioning of the target object is the prerequisite for its successful grasp of the target object. However, general intelligent robots cannot easily identify and locate the target object in a complex environment like humans, usually through image processing. Algorithms are used to extract object features for recognition, but it has always been a difficult problem to accurately and quickly recognize targets in complex environments

[0003] In addition, the traditional robotic arm usually grasps the target according to the established process, and cannot accept external information. When the target position changes, the control program needs to be reset, which affects the efficiency of the robotic arm.

At the same time, the robotic arm does not have the function of autonomous grasping

Images