Tak-Kuen John Koo, Graduate Student
Frank Hoffmann, Postdoctoral Researcher
(Professor S. Shankar Sastry)
Funding Sources: ARO-MURI DAAH04-96-1-0341,
The research project is concerned with the design of a hybrid control system[1] for an unmanned helicopter. The proposed control scheme uses discrete-time automaton in the upper level to handle discrete events, such as execution of external commands, safety issue, switching between flight modes. A Fuzzy Logic Controller(FLC) is employed for the continous-time lower level control, which maps perceived sensor information to control action based on the context provided from the upper level.
Fuzzy Logic(FL)[2] offers a methodology for reasoning and computing that respects the human perceptual capabilities, because knowledge is described by means of linguistic variables and rules in place of numerical data. FLC provides the designer with a conceptual framework, which allows him to understand, manipulate, and organize the control behaviour without requiring a precise mathematical model of the process. An additional goal of the project is to develop a Fuzzy Linguistic Model(FLM)[3] of the helicopter, which will be used for model-based control design. To enhance the performance of the FLC, we propose the integration of Genetic Algorithms(GA) with FLC[4] in order to adapt the controller parameters with regard to performance criteria and to the variations in system parameters and environment.
(No figures supplied)