Application of Fuzzy Control to Helicopter Navigation Carla Cavalcante*, Janette Cardoso*, Josué J. G. Ramos**, Othon R. Neves** * LCMI-EEL-UFSC, CP 476, 88040-900 Florianópolis (SC), Brasil ** CTI/Gyron Tecnologia, Florianópolis (SC), Brasil e-mail carla@lcmi.ufsc.br - janete@lcmi.ufsc.br Abstract—Fuzzy control has shown to be useful in handling non-linear systems and ill-defined or imprecise problems that depend on the operator skill, as is the case of helicopters. This paper presents a simulated environment application of fuzzy control to a helicopter navigation system. This work corresponds to a suitability analysis for the implementation of this system on board of an unmanned helicopter developed by Gyron Tecnologia. The system organization is described, along with the knowledge base controller design. The results obtained for hover flights are shown. I. Introduction This paper shows a fuzzy control application to helicopter navigation system. A helicopter is an intrinsic complex, non- linear, unstable process, with coupled modes. Several control techniques have been used by designers aiming to control this kind of aircraft, generally using linearized models ([FGF76], [Apk89], [CP72], [Tak93]). Our approach consists in translating the linguistic control strategy described by a helicopter pilot into an automatic control strategy. This qualitative approach allows to consider the pilot expertise on conducting unstable aircrafts, making it possible to the system to execute the pilot actions, as take off, landing and hover. In fact, applications of fuzzy control ([Lee90], [SK92], [SGB93], [MS85]) have indicated effective utilization of fuzzy control in the context of complex ill-defined processes that can be controlled by skillful human operators, as the helicopter. The basic configuration of a fuzzy logic controller comprises four components [Lee90]: • a fuzzyfication interface, which performs a scale mapping on the range of values of input variables into corresponding universes of discourse; • a knowledge base, which consists of a data base and a linguistic fuzzy control rule base; • a decision-making logic, which simulates human decision- making; • a defuzzyfication interface, which produces a non-fuzzy control action from an inferred fuzzy control action. In this work, we focus on the system navigation structure and knowledge base design, describing the choice of variables, sets and rules. This paper is organized as follows. Section two introduces the helicopter, its control and movements, as well as the control problems to be considered while designing the navigation system. The system navigation design, the tasks it should execute and its hierarchical structure (the mission interpreter, the task level and the fuzzy controller) are described in section three. Section four presents the fuzzy controller and the choice of input and output variables, fuzzy sets and fuzzy rules. Section five shows the simulation results for hover flight, with comments about the tuning phase. Finally, the conclusion and future work are presented in section six. 1 Figure 1: Helicopter aerodynamics components [Pall83]