Open AccessApplication of the Q-learning algorithm for the intellectual mode of the manipulator arm for moving to a present position
Author(s) -
A. Yu. Yakovlev,
A. A. Krasnaya,
Sergey Medvedev
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1902/1/012138
Subject(s) - robotic arm , position (finance) , kinematics , manipulator (device) , computer science , movement (music) , algorithm , mode (computer interface) , robot manipulator , artificial intelligence , q learning , control theory (sociology) , robot , simulation , control (management) , reinforcement learning , physics , finance , classical mechanics , economics , philosophy , operating system , aesthetics
The article is devoted to the search for a certain kinematic scheme that determines the movement of the arm links of an industrial manipulator. A robotic complex with a single-circuit hydraulic system of arm element drives is considered. The problem of the automatic transition of the arm to a user-specified state has been solved, taking into account the restrictions on the simultaneous movement of the arm links and the minimum and maximum heights of the manipulator. To solve the problem, we used the machine learning method – Q-learning.