Design and implementation of two-wheeled intelligent robot
| Autor: | Tong-Ying Chiang, 江東穎 |
|---|---|
| Rok vydání: | 2009 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 98 This thesis is divided into two parts, the first part includes the derivations of control equations. In the control scheme, the fuzzy inference is designated as a main controller and the neural network is designated as auxiliary part. The neural uncertainty observer is added to fuzzy balance controllers. It is utilized to observe the external load disturbance, nonlinear friction and unstructured uncertainties, etc.. According to the observed result the control difficulty of balancing controller can be reduced. To verify the effectiveness of the proposed control architecture. The MATLAB simulation is carried out. The second part is the experimental implementation of Parallax Inc’s Boe-Bot robot. The Accelerometer 2125 and ADXRS610 are used to measure the inclination angle and the angular velocity of Boe-Bot robot. The inclination angle and the angular velocity are two inputs of balance controller. The conventional PD, fuzzy and the proposed neural fuzzy control algorithms are programmed the BS2 module, respectively. The control signals are evaluated by the PIC16C57 and sent to the left and right motors. The data stored in BS2 and the waveform in the oscilloscope are acquired to observe the balancing condition. After watching the condition of balance and amended algorithms. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|