Improvement of robot navigation using fuzzy method

摘要

In this paper a technique for autonomous navigation of mobile robots is presented. The most important advantage of this method is to ignore the physical model of the robot and that the robot model is considered as an unknown but predictable system. This approach can be executed in environments which robot navigation is done using global sensors e.g. a camera installed in the environment to sense and control the robot and also every feedback system which is able to measure the velocity of the robot can be used for this technique. In this approach, the well-known fuzzy method, Takagi-Sugeno, has been applied to estimate the dynamic model of robot. Our technique has been applied successfully in small-size soccer robots league for improvement of soccer robots navigation. In this league a shared vision system is used to obtain the position and orientation of the playing robots and position of the ball and no interference is performed by human during a game. The fuzzy system produces the desired model between inputs and outputs of navigation system with less process time and more accuracy rather than other available methods.