Research on robot arm control based on Unity3D machine learning. (September 2020)

• Record Type: Journal Article
• Authors: Lin, Mu; Shan, Lijun; Zhang, Yongchun
• Is Part Of: Journal of physics. Volume 1633(2020)
• Publication Date: 2020-09
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1633/1/012007
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: Based on the Unity3D engine, the article uses deep reinforcement learning strategies to train the robotic arm through the reward function, and realizes machine learning and intelligent control of the robotic arm. After training and learning, the robotic arm can quickly and accurately find movement point in the environment and has high environmental adaptability. The application of powerful deep reinforcement learning strategies and virtual reality technology in engineering technology teaching has improved the teaching effect and teaching efficiency of mechanical structure cognition, curriculum design and other teaching links.