Paper Info
Title
Adaptive neural network controller of an intelligent commode robot for trajectory tracking
Abstract
An intelligent commode robot with omnidirectional mobility has been developed for the elderly and the disable to improve their life quality. Considering a load change and a shift of the center of gravity due to different users, an adaptive controller based on radial basis function neural network (RBFNN) is designed to track a desired trajectory of the commode robot system. The unknown parameters in dynamics of the system are estimated on-line by the RBFNN. The e1-modification algorithms are used to tune the weights of the RBFNN and a learning rate dependent on time is designed to update the weights to accelerate the training rate of the neural network. Furthermore, a sliding mode control term is selected to compensate estimation errors of the dynamic function and unknown bounded disturbances. For the bounded predefined trajectory, the uniformly ultimately bounded stability of the tracking errors and neural network weight errors has been demonstrated based on the Lyapunov theory. Simulation results showed that the proposed controller can precisely follow the desired trajectory with the load change and the shift of the center of gravity, which will greatly improve the reliability and safety of the commode robot in movement.
Year
DOI
Venue
2017
10.1109/CBS.2017.8266121
2017 IEEE International Conference on Cyborg and Bionic Systems (CBS)
Keywords
Field
DocType
Commode Robot,adaptive control,radial basis function neural network,load change,center of gravity shift
Lyapunov function,Control theory,Adaptive system,Control theory,Computer science,Robot kinematics,Artificial neural network,Trajectory,Mobile robot,Sliding mode control
Conference
ISBN
Citations 
PageRank 
978-1-5386-3195-9
0
0.34
References 
Authors
1
4
Name
Order
Citations
PageRank
Jia Ma11074.19
Huaxin Liu256.38
Shuoyu Wang38927.69
Qiang Huang426691.95