Study of Neuron Sliding Mode Position Control Based on CMAC

2010 ◽  
Vol 439-440 ◽  
pp. 440-444
Author(s):  
Xiao Juan Shi

Combining cerebella model articulation controller with sliding mode control, a novel control algorithm is designed. It adopts neuron PID controller as the teacher signal of CMAC. The control error rapidly tends to zero by adjusting weight of each corresponding unit in CMAC storage. The control algorithm eliminates the chattering in the conventional sliding mode control. It also has high real-time, strong robustness and better convergence. The control algorithm is applied on position servo system of NC machining tool. The experimental results show that it improves dynamic performance and static performance of position servo system. It can realize rapid and exact position servo control of NC machine tool.

2011 ◽  
Vol 317-319 ◽  
pp. 1490-1494 ◽  
Author(s):  
Bao Quan Jin ◽  
Yan Kun Wang ◽  
Ya Li Ma

The parameters uncertainty and external disturbance play a negative role to improve electro-hydraulic position servo system performance. The valve controlled cylinder system model is established, using the traditional PID control strategy and reaching law control strategy for simulating the system, respectively, the two methods have similar control effects in the ideal model, but considering the external disturbances, the index approaches sliding mode control law has better response speed and stability. Research shown that sliding mode control algorithm has an important role for improving the performance of hydraulic servo position control system.


Author(s):  
Shan Jiang ◽  
Wenhao Feng ◽  
Peng Zheng ◽  
Jun Liu ◽  
Jun Yang

This paper presents a control methodology that enables nonlinear model-base control of pulse width modulated (PWM) pneumatic position servo control system of a magnetic resonance (MR) compatible needle insertion robot used for MRI-guided prostate brachytherapy. Sliding mode control with a boundary layer is investigated based on the equivalent continuous-time dynamic model. Considering chattering in sliding mode control applications, especially in motion control of the robot, a fuzzy method is developed to tune the boundary layer of the sliding mode controller. Simulation and experimental results indicate that the controller is effective in both position control and trajectory tracking. With the boundary layer increasing, the steady state error of the response is becoming larger; however the maximum error is less than 0.2 mm. This steady state performance meets the accuracy requirement.


Sign in / Sign up

Export Citation Format

Share Document