Design of the neuro-like learning control system for a vehicle

The paper is devoted to design of the neuro-like learning control system for a vehicle or an autonomous mobile robot with a use of artificial neuro-fuzzy networks advantages. Operation of the vehicle control system is aimed to solve the obstacle avoidance task in uncertainty conditions. We described the advantages of artificial intelligence methods in the tasks of control systems design using neuro-fuzzy approach. The structure of learning control system consisting of three modules (motion direction control module; velocity control module; obstacle identification module) is given. The stages of each module design are defined and the structures of neuro-fuzzy networks for the implementation of modules are chosen. We presented description of the operation algorithms of the motion direction control module, the velocity control module, the identification module for the obstacles on vehicle path, and the structure of neuro-like learning control system. Key-Words: control, uncertainty, training, neuro-fuzzy networks, vehicle, autonomous mobile robot.