Motion Control of an AUV Using Neural Network Adaptive Controller

A0050 Ji-Hong Lee Korea Research Institute of Ships and Ocean Engineering (KIRSO), KORDI

A0051 Pan-Mook Lee Korea Research Institute of Ships and Ocean Engineering (KIRSO), KORDI

A0052 Sang-Jang Lee Department of Electronics Engineering, Chungnam National University

As the use of autonomous underwater vehicles (AUVs) increases in a

diversity of undersea applications, the control system has become one

of the most critical subsystems of the vehicles. The dynamics of AUVs

are highly nonlinear and their hydrodynamic coefficients vary with

different operational conditions, so it is necessary for the high

performance control system of an AUV to have the capacities of

learning and adapting to the change of the AUVs dynamics. There are

many methods to compensate the uncertainties of underwater vehicles,

such as neural network based learning controllers, nonlinear adaptive

controllers, sliding mode controllers, etc.

This paper presents a neural network based nonlinear adaptive

controller for an AUV. The presented controller consists of three

parallel schemes; a linear feedback controller, a sliding mode

controller and a linearly parameterized neural network (LPNN). The

linearly parameterized neural network is used to approximate the

nonlinear uncertainties of the AUVs dynamics, and the sliding mode

controller is introduced to compensate the reconstruction errors of

the neural network and the disturbances of the AUVs dynamics.

The initial values of the neural weights can be set at zeros and

adapted with a learning algorithm, so the neural network requires no

off-line evaluation. A basis function vector of the neural network is

selected according to the physical properties of the underwater

vehicle. Lyapunov theory is used to guarantee the stability of the

proposed controller in which the estimation errors of the neural

networks weights converge asymptotically and the trajectory tracking

errors converge to zeros asymptotically.

Numerical simulations for the motion control of an AUV are performed

to illustrate the effectiveness of the proposed controller. Through

the simulation studies, it can be concluded that the presented

controller is suitable for the high performance motion control of the