Projeto e analise de controladores nebulosos e sua aplicação para controle de juntas roboticas

AUTOR(ES)
DATA DE PUBLICAÇÃO

1997

RESUMO

This work presents a methodology for analysis and design of a class of fuzzy controllers and it contributes to formally justify their advantages. The methodology is constituted by methods for linguistic control rule designing and for qualitative and quantitative studies that provide stability analysis and performance criteria. The approach used for rule base design is based on the qualitative information described by the step response of a linear and underdamped processo Moreover, some metarules that relate the basic experience of a human operator are also used. As a result of the previous approach, the qualitative analysis is concerned with the qualitative study of the rules and it is based in a phase plane of linguistic states. This analysis allows to point out the role of each rule on the control system performance and behaviour and it also contributes to rule modification. The approach proposed for quantitative analysis introduces an alternative way to get the mathematical description of the linguistic rules. As a result, it provides a rigorous study of the fuzzy controller and therefore it open new study perspectives for stability analysis. A procedure for performance criteria design is presented. Besides this, it is demonstrated that classical controllers of PID type are specific functions of fuzzy controllers. The concepts proposed in this thesis are applied to the development of a fuzzy controller to joint control of a robotic manipulator. The simulation results enphasize the advantages and the better performance of this controller when compared with classical controllers of PID type

ASSUNTO(S)

robotica sistemas difusos controle de processo

ACESSO AO ARTIGO

Documentos Relacionados