The methods for the detection of faults of actuators, devices and algorithms for fault tolerance were considered. The problem of fault diagnosis is an important aspect in the design and use of control systems. Marine objects are complex hardware and software systems involved in the tasks taking place in uncertain environments, there is a way the degree of autonomy becomes an important issue. Therefore, the ability to detect and to continue to maintain the health while an error or equipment failure is a key factor for the successful completion of the mission or the resumption of the previous state of the object.

Keywords: Modeling, control systems, diagnostics, fault tolerance, neural networks, AUV

The paper considers the use of ground station operator’s software that provides the diagnosis of automatic control system of marine vehicles, based on the monitoring of the functional state of the system elements with depth to the type of failure. Also considers the use of neural networks for solving the problems of monitoring, control and state diagnostics of the actuators and sensors. Describes the principles applying of active fault tolerance and flexible functionality recovery of marine mobile vehicle’s control system

Keywords: Modeling, control systems, diagnostics, fault tolerance, neural networks, AUV