Effect of Fuzzification and Defuzzification Methods on the Behavior of an Autonomous Mobile Robot Navigation

M.A. Eshtawie and A.M. Salah (Libya)


Fuzzy control, Mobile Robot, Fuzzification, Defuzzification, Navigation, Sensor Readings


For an autonomous mobile robot (AMR) to be useful in realistic situation, it must be able to safely interact with other objects whether they are other mobile robots or human beings. On the other hand, Controlling mobile robot navigation problem is a highly non-linear and complicated task that needs both human experts thinking and numerical values obtained from sensors mounted on the body of the robot. One major feature of fuzzy logic is its ability to express the amount of ambiguity in human thinking in a comparatively undistorted manner. Therefore, during the past several years fuzzy logic control has emerged as one of the most active and fruitful areas for research in the application of intelligent system design. This paper presents the effect of different fuzzification and defuzzification methods on the response of the FLC and therefore, on the behavior of an autonomous mobile robot moving in unstructured environment. The FLC is constructed using a software program written in C++ programming language. In this paper, three different fuzzifiers have been used for inputs to be interpreted and compared to rules in the rules base. These fuzzifiers are the triangular, trapezoidal and Gaussian. On the other hand, the center of gravity and center average were used as defuzzification methods to get the results of the fuzzy controller.

Important Links:

Go Back