Fusion of infrared sensors and camera for mobile robot navigation system - simulation scenario

Abstract

This paper presents a simulation scenario of a mobile robot navigation system. Navigation system is based on data collecting from three infrared sensors and camera, on which basis the fusion process of detection and avoidance of obstacles is realized. Information from camera was used to detect edges of the obstacles in the environment, while infrared sensors were used to measure the distance from the obstacles. Multilayer perceptron network, trained with backpropagation algorithm, was used for classification of detected obstacles. Experiment was realized through simulation in the simulation environment “Robotino SIM”. The control algorithm was implemented in MATLAB. Sensor fusion has proven to be a much better solution than using infrared sensors or cameras separately. This experiment showed that the developed algorithm gives very good results (accuracy: 89.61%), and the navigation system itself performs required tasks of detecting and avoiding most of the obstacles on which it was tested.