Title :
Reinforcement learning control for a robotic manipulator with unknown deadzone
Author :
Yanan Li ; Shengtao Xiao ; Shuzhi Sam Ge
Author_Institution :
Inst. for Infocomm Res., Agency for Sci., Technol. & Res., Singapore, Singapore
Abstract :
In this paper, an actor critic neural network control is developed for a robotic manipulator. Both system uncertainties and unknown deadzone are considered in the tracking control design. Stability of the closed-loop system is analyzed via the Lyapunov´s direct method. The critic neural network is used to estimate the cost-to-go and the actor neural network is used to make the cost-to-go converge. Simulation studies are conducted to examine the effectiveness of the proposed actor critic neural network control.
Keywords :
Lyapunov methods; closed loop systems; control system synthesis; learning systems; manipulators; neurocontrollers; stability; uncertain systems; Lyapunov direct method; actor critic neural network control; closed-loop system; cost-to-go estimation; reinforcement learning control; robotic manipulator; stability; system uncertainties; tracking control design; unknown deadzone; Artificial neural networks; Learning (artificial intelligence); Manipulator dynamics; Vectors; Reinforcement learning; deadzone; neural networks; robot control;
Conference_Titel :
Intelligent Control and Automation (WCICA), 2014 11th World Congress on
DOI :
10.1109/WCICA.2014.7052780
