SMC-fat based adaptive control of robot manipulator unknown time varying uncertainties /
The controlling of robotic arm is really challenging due to the involvement of various uncertainties including time varying payload, disturbances and friction. These challenges attract many researchers to develop advanced control strategies for robot arm. However, most of the developed adaptive cont...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Gombak, Selangor :
Kulliyyah of Engineering, International Islamic University Malaysia,
2016
|
| Subjects: | |
| Online Access: | Click here to view 1st 24 pages of the thesis. Members can view fulltext at the specified PCs in the library. |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The controlling of robotic arm is really challenging due to the involvement of various uncertainties including time varying payload, disturbances and friction. These challenges attract many researchers to develop advanced control strategies for robot arm. However, most of the developed adaptive controllers focus on time invariant uncertainties or need the utilization of regressor for time varying uncertainties. This research represents the formulation of a new Sliding Mode Control- Function Approximation Technique (SMC-FAT) based adaptive controller for a robot manipulator carrying unknown time-varying payload with the presence of time-varying disturbance and friction. The limitation of the previous controllers to cope up with wide range time-varying uncertainty is solved by using FAT expression where FAT represents the uncertainties. In these methods, firstly Taylor series is used as a basis function. The stability of the controller can be proved and using proper Lyapunov function, the update law can be derived easily. Regressor matrix is not necessary in the proposed control law. Then neural network is utilized to reduce the number of basis function and decrease computational time. Simulation test and hardware experimental tests have been conducted to verify the effectiveness of the controller. Three different time-varying uncertainties in sinusoidal, sawtooth and random functions have been considered as the payload and disturbance in the computer simulation. The results show that the controller has successfully compensate the time-varying uncertainties and disturbances with a maximum error less than 0.002 rad. Low number of basis function is necessary in the Radial Basis Function Neural Network (RBFNN) compared to Taylor series basis function. Furthermore, 3 Degree of freedom (DOF) robot manipulator hardware has been developed for the experimental test. The results validates the controller performance in both Taylor series and RBFNN based basis function, where the maximum experimental error is 0.0427 percent. |
|---|---|
| Physical Description: | xviii, 113 leaves : ill. ; 30cm. |
| Bibliography: | Includes bibliographical references (leaves 98-102). |