Experimental Evaluation of Fuzzy Logic Control of a Flexible Arm Manipulator
AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH
Pagination or Media Count:
The purpose of the research reported in this thesis was to explore fuzzy logic control of a single link, flexible robot arm. The performance and stability robustness properties of fuzzy logic control were evaluated in simulation and with hardware in the loop in the University of Washington Control Systems Laboratory. The plant model and the Integrated System, Inc., AC100-C30 real time control prototyping system were first validated using a PID controller, which was also used as a comparison for the performance of various fuzzy logic controllers. The control objective was to slew the arm rapidly to commanded positions while minimizing the excitation of the links flexible modes. The available sensor was a hub angle sensor no tip position or tip rate information was used for the controllers. The stability robustness of the PID controller and the fuzzy logic controller were evaluated in hardware for mismodeled system parameters. Tests were conducted for increased tip mass relative to the design condition controller as well as for a significantly more flexible link than the design condition. The fuzzy controller was found to be stable for all test conditions, while the PID controller was unstable for one of the off-design conditions.