Information system for automatic orientation in 2D space of mobile robot

Abstract

In the modern intelligent robotic systems, the most common problem is that the development methods, techniques and technological solutions for autonomous navigation of robots. Under the navigation, orientation should be understood smart device in space. This article describes a way to autonomous robot movement in the room a hallway-type room on a horizontal surface within the same floor. The solution of this problem is achieved by performing three processes: selection of areas, conservation areas, and target detection. Unlike existing solutions, we propose an improved method for the orientation of the object in a horizontal plane based on using neural networks as base classifier data. The proposed method makes it possible to install four ultrasonic sensors on the robot body with a further detection of the distance to objects and walls, as well as remembering, recognition and forecasting of the route. Functionally, this method allows a certain accuracy, as well as the measurement errors, determine the coordinates of the robotic device to the smart card on the horizontal surface area without the use of global positioning systems GPS and GLONASS, and the cellular GSM network.