Trajectory based fuzzy controller for indoor navigation

The experimental results of fuzzy based indoor navigation system are presented in this article. The system was built with simplicity in mind, employing roof mounted visual sensors and fuzzy logic for the guidance and positioning of an autonomous mobility device with real-life noisy and delayed visual data processing. The fuzzy control strategy presented bellow works on a given (predefined) trajectory principle. The position of the device in the 2D plane, the distance from the given trajectory, orientation and control tasks are evaluated according to visual data. The controller controls the angular velocities of two device drive wheels based on the distance to the segment of the given trajectory and the orientation. Paper presents the control model and algorithm of a real-life prototype, as well as the quite promising experimental evaluation of the indoor autonomous navigation capabilities.