Problems of a trajectory planning in autonomous navigation systems based on technical vision and AI

Problems of realization of completely autonomous systems of navigation are considered in a set of spheres of human activity today. The most important task is today the attempt to create on the basis of system of technical sight completely autonomous system of navigation for motor transport, the marine transport and aircraft. In article the problem of creation of such system for the drone which is carrying out research or rescue operations on the difficult area is considered. And it is concrete in the wood. The main objective of system is finding of footpaths among trees on which the drone can follow. The main tool of the solution of this problem offers use of artificial neural networks of deep training or convolution networks. The quantity of layers and dimensions of local networks at their use is proved in problems of autonomous navigation of drones on the basis of systems of technical sight.