Grabbing action control method and system for a highly intelligent robot

Abstract

The invention discloses a grasping action control method and system of a highly intelligent robot, wherein the control system includes an environment perception module, an identification module, a judgment module, an action planning module and an execution module. Through this technical solution, the autonomous grasping action planning control of highly intelligent robots can be realized, and the effective grasping of various objects can be realized; for some small objects that are fragmented and easy to move with the environment, accurate grasping can be achieved through iterative planning. Tracking and grasping can also realize the transportation of grasped objects; and realize intelligent autonomous recognition and autonomous planning through pre-artificial intelligence training in recognition and action planning, without the need for users to input corresponding recognition models and action models, greatly improving The intelligence of the robot using this kind of grasping control system is improved.

Description

technical field [0001] The invention relates to the technical field of high-intelligence robot control, in particular to a high-intelligence grasping action control method and system. Background technique [0002] With the rapid development of science and technology, highly intelligent robots have been widely used in various fields such as industrial production, logistics, metallurgy and medical treatment, special disaster relief, 3C industry, etc., and can replace manpower for difficult tasks such as grasping and handling, thereby significantly improving work. At the same time, it can replace manpower to carry out some complicated and dangerous work to ensure the personal safety of operators. [0003] At present, the commonly used method is to control the robot to grab the workpiece through the guide rod or manual operation, or to locate the workpiece through the visual structure and then plan according to the preset motion model, which is difficult to achieve universal and...

AI Technical Summary

Problems solved by technology

[0003] At present, the commonly used method is to control the robot to grab the workpiece through the guide rod or manual operation, or to locate the workpiece through the visual structure and then plan according to the preset motion model, which is difficult to achieve universal and high-precision Realization, there is an urgent need for a highly intelligent robot motion planning control method to solve the above practical application problems

Images