Proposed system of artificial Neural Network for positioning and navigation of UAV-UGV

Abstract

This paper deals with a path planning and intelligent control of Belarus Mobile Robot which should move safely in unstructured environment by means of Neural Network System (NNS). The approach combines input variables from different resources such as: GPS, RVS (Robot Vision System), and QVS (Quad-copter Vision System). A model based on competitive learning has been developed for robot motion planning. Our method of the construction of a collision-free path for moving robot among obstacles is based on neural networks. A new algorithm for updating the neural activities has been proposed to solve the positioning and navigation problems for mobile robot.