Mechanical arm six-degree-of-freedom visual closed-loop grabbing method based on TSDF three-dimensional reconstruction

Abstract

The invention relates to a mechanical arm six-degree-of-freedom visual closed-loop grabbing method based on TSDF three-dimensional reconstruction. Comprising the following steps that 1, a mechanical arm base coordinate system and a camera coordinate system are calibrated through a Zhang Zhengyou camera calibration method and an ArUco Markers calibration method; 2, performing three-dimensional scene reconstruction on the acquired image information by using a TSDF function so as to reduce environmental noise points between objects; 3, establishing a reinforcement learning network model; 4, the predicted grabbing posture of the end effector is reversely projected into the three-dimensional reconstruction scene, and the current predicted grabbing quality is judged; 5, mechanical arm grabbing movement is completed through forward and inverse kinematics of the robot; sixthly, reinforcement learning model training is conducted, so that the mechanical arm completes the grabbing action; according to the mechanical arm six-degree-of-freedom visual closed-loop grabbing system based on TSDF three-dimensional reconstruction, the defects in the prior art are overcome, the mechanical arm six-degree-of-freedom visual closed-loop grabbing system is easy to implement, high in applicability and based on TSDF three-dimensional reconstruction, environment noise caused by shielding and stacking of objects is reduced, and depth errors caused by interference of a single visual sensor are reduced. In addition, the system can realize rapid and real-time target detection and complete the grabbing action while ensuring high precision.

Description

technical field [0001] The invention belongs to a technology of grasping objects with six degrees of freedom of a mechanical arm in a non-structural environment, in particular to the field of robot grasping control. Background technique [0002] Object grasping has always been an important problem in robotics, but there has never been a satisfactory corresponding solution. Thanks to the multi-functional manipulation system of the robotic arm, it can flexibly control the end effector with a high degree of freedom in three-dimensional space, realize the dexterous grasping of objects and the dynamic response ability to environmental changes. Recently, with the rapid development of deep learning and reinforcement learning and the construction of corresponding systems, various feasible ideas have been provided for the intelligent grasping method of robotic arms. [0003] Although robotic arm 6-DOF grasping control has high practical value for more complex operating environments,...

AI Technical Summary

Problems solved by technology

However, the success rate of this algorithm for unrecorded target objects and object geometry in the data set will be relatively low, so it cannot be extended to new application scenarios

Images