Deep reinforcement learning robot control method based on priority experience playback

Abstract

The invention discloses a deep reinforcement learning control algorithm based on a priority experience playback mechanism. The priority is calculated through employing state information of an object operated by a robot, an end-to-end robot control model is completed through employing a deep reinforcement learning method, and a deep reinforcement learning intelligent agent is enabled to autonomously learn and complete a specified task in the environment. In a training process, the state information of the target object is collected in real time and used for calculating the priority of experience playback, and then data in an experience playback pool is sampled and learned by a reinforcement learning algorithm according to the priority to obtain the control model. According to the method, onthe premise of ensuring the robustness of the deep reinforcement learning algorithm, environment information is utilized to the maximum extent, the effect of the control model is improved, and the learning convergence speed is increased.

Description

technical field [0001] The invention belongs to the field of robot control, and in particular relates to a robot control method based on a virtual environment, deep reinforcement learning, and a priority experience playback algorithm based on object position changes. Background technique [0002] At present, most robot space grasping technologies mainly pre-set the possible behaviors of robots or are based on traditional 3D vision algorithms. However, with the continuous expansion of robot application fields, the tasks faced by robots are becoming more and more complex. When faced with complex tasks, complex visual calibration methods and visual modeling methods are required, and designers cannot make effective predictions for rapidly changing environments, making it difficult to make reasonable predictions about robot behavior. [0003] Reinforcement learning is an important learning method in the field of machine learning. In the application of the field of robotics, the ...

AI Technical Summary

Problems solved by technology

However, there are great difficulties in scientific and effective priority design. Currently, there is a lack of a priority design method for robot arm manipulation tasks.

Images