An enhancement in conventional potential field using a fuzzy system for navigation of a sumo robot

An enhancement in conventional potential field is presented in this paper for autonomous sumo robot class navigation with dynamic environments including moving target and static obstacle. This method determines the velocity and the path taken by the robot in order to reach your target in the shortest time possible. This improvement on the conventional potential field is used to determine the velocity and position of an autonomous sumo robot class, which is intended to push the opponent out of the arena. This hybrid method is composed of potential field approach with fuzzy logic, and in this study the Mamdani model is used to obtain the necessary variables to determine the velocity and direction of the robot. Two fuzzy systems are used, both with three inputs and one output. The only difference between the two fuzzy systems is the number of linguistics terms. One simulation has been done in the MATLAB to verify the enhancement of the hybrid method in relation to the conventional potential field. The simulation represents the trajectory of the robot until find the target. Accordingly with the simulation done, the hybrid method was 14% more effective.