Adamiv O. Models and intellectual means for adaptive autonomous mobile robot control.

This dissertation was devoted to improvement of existing and development of new models and intellectual means for autonomous mobile robot (AMR) control for providing of navigation tasks in the complex nonstructured environments. The regular iteration method of gradient search with local adaptive measurement of limitations of second class is developed in the thesis. It provides the reaching to the goal of AMR movement in the complex nonstructured environment. In order to control the trajectory of AMR movement the methods for maneuvers providing was improved. It provides the increasing of probability of maneuvers executing up to 45 percent using the formation of adaptive training set for artificial neural network. In order to control the AMR in the dynamical environments the intellectual robot control model was improved. This allowed to measure the main data flows and time delays between functioning modules of AMR. As a result the productivity of software and hardware AMR components for navigation taskswas increased.