Robot stirring and grabbing combination method based on deep reinforcement learning

Abstract

The invention discloses a robot stirring and grabbing combination method based on deep reinforcement learning. The robot stirring and grabbing combination method comprises the following steps: building a robot stirring and grabbing execution platform in a real environment and a robot stirring and grabbing execution platform in a simulation environment respectively; modeling a stirring and grabbingprocess of the robot into a Markov process, and constructing a state space, an action space and a reward function; constructing a robot stirring and grabbing learning framework based on deep reinforcement learning and a deep reinforcement learning network; then, repeatedly carrying out stirring and grabbing action attempts on the simulation platform, collecting experience sample data, and carrying out self-supervised training on the deep reinforcement learning network according to the Markov process; and finally, migrating the trained network model to a real environment, and carrying out actual training on the robot in the real environment. When facing a random grabbing scene and a novel grabbing object, the robot can sense and analyze the scene and decide to execute a stirring or grabbing action, an action position and an action direction.

Description

technical field [0001] The invention relates to a robot stirring-grabbing combination method based on deep reinforcement learning, which belongs to the technical field of robot application. Background technique [0002] Grasping is one of the basic operations of robots, and it is also the basis and key for robots to complete various tasks. As my country gradually enters a period of rapid development of robot technology, robot applications are gradually moving from traditional industrial fields to home services, warehousing and logistics, etc. Grabbing simple types of objects in fixed scenes is far from meeting the grasping needs. In the face of unstructured grasping environment and unknown objects, the robot needs to have stronger adaptability and the ability to actively change the environment to reduce the complexity of the scene and improve the success rate of grasping. [0003] The deep learning method can learn and mine feature representations better than manual design ...

AI Technical Summary

Problems solved by technology

[0005] Purpose of the invention: In view of the shortcomings of the existing grasping methods and the problem that there is not enough grasping space when the objects are arranged closely, the present invention proposes a combined grasping method based on deep reinforcement learning of robot agitation and grasping , using self-supervised learning to enable the robot to acquire the ability to stir and grasp without the need for a target model and a manually labeled dataset

Images