Observer-based neural adaptive formation control of autonomous surface vessels with limited torque

• A neural adaptive formation controller is proposed for surface vessels.
• A new formation model is proposed that inherits all properties of vessel dynamics.
• The controller reduces the risk of actuators saturation by saturation functions.
• Neural adaptive robust techniques compensate disturbances and unmodeled dynamics.
• The controller does not require velocity signals by using a velocity observer.

In this paper, the output feedback formation control of autonomous marine surface vessels with limited torque input is addressed. In order to successfully design a formation control system, a second order formation dynamic model is developed based on the leader-following strategy. The controller is designed by employing generalized saturation functions in order to reduce the risk of actuator saturation. A nonlinear saturated observer is also introduced to estimate velocities of the followers. Multi-layer neural network and adaptive robust techniques are also incorporated in the design of the control system to preserve its robustness against uncertain nonlinearities and environmental disturbances which are induced by waves and ocean currents. Lyapunov’s direct method is used to show that all signals of the closed-loop system are bounded and tracking errors are semi-globally uniformly ultimately bounded. Finally, computer simulation results are provided to demonstrate the efficacy of the proposed formation controller for a number of surface vessels.