Goal-Driving Control as a Base Model of the Feeling Artificial Intelligence

When creating unmanned systems (US), the main attention is paid to the problem of autonomy. The use of artificial intelligence (AI) is one of the ways to increase the level of US autonomy in a disordered environment. Today, a new generation of Feeling AI (FAI) is aimed to support the process of control of the implementation of action plan of Autonomous Intelligence US (AIUS). The peculiarity of the AIUS control task is that making decision in real time about current action use the state of the plan's implementation, the current situation and the ability to implement the remaining part of the action plan to achieve the goal. The article examines sequence control and rules-based control methods. The structure of a multi-layer distributed fuzzy logic system (FLS) combined with production rules system is given. A modified fuzzy inference engine which thanks to the introduction of a fuzzy certainty factor as a complex number is able to uniformly process both linguistic variables of FLS and facts from production rules system, was considered. An example of AIUS control task and computer experiments with a wheeled robot are given.