Soft Computing Based Control Algorithm on AUV

A0014 H. H. Chen Shenyang Institute of Automation, Chinese Academy of Sciences

A0012 Z. W. Xing Shenyang Institute of Automation, Chinese Academy of Sciences

A0025 Y. Zhang Shenyang Institute of Automation, Chinese Academy of Sciences

A0013 X. S. Feng Shenyang Institute of Automation, Chinese Academy of Sciences

Currently, most of AUVs are survey research vehicles, and have

performed in ocean environment surveys, oceanographic sections and

hydrographic samples, subsea resources surveys. However, AUVs are

limited by some key technologies such as energy, acoustic

communications, sensors, navigation and positioning, control and

artificial intelligence, it is difficult for to AUVs perform

complicated subsea tasks such as pipeline survey, subsea oil field

inspection and intervention, undewater connection and recovery. AUVs

of highly accuracy dynamical position and target tracking is one of

the keys of performing above tasks. Due to the poor control precision

of common control method, a new control method must be adapted.

Affected by the hydrodynamics and the environment disturbance,

dynamics of AUVs are highly nonlinear, multi-freedom coupled, and time

varying. Due to the uncertainty of the hydrodynamics and the

complexity of ocean environment, it is difficult to establish highly

accurate dynamics of AUVs. Usually, approximate hydrodynamics model

of AUVs is obtained by experience and experiment.

The environmental disturbance of the ocean includes wave, ocean

current and imbalance force . The imbalance force is brought from the

imbalance of vehicle structure and hydrodynamics, the imbalance force

model and the ocean current model both are 1st-order and showly-vary

ing disturbance. Wave forces can be divided into 1st-order wave

disturbance and 2st-order wave drift force. In this paper, we

supposed the environmental disturbance of the ocean as the combination

of fundamental harmonic, 1st-order harmonic and 2nd-order harmonic.

Usually, we use singleton to fuzzize input variable of fuzzy logic

controller, but in the situation with serious disturbance, this

traditional fuzzy logic controller can't work well.

Considering the problems mentioned above, a novel control algorithm

based on Soft Computing is proposed in this paper. The main idea of

the algorithm is that, firstly, we fuzzize input by Gauss function,

and choose Gauss function as membership function, and choose center-

average defuzzier, then we can build a neural-net based on fuzzy logic

control, and some optimization method has been applied to the neural-

net. By studying we draw the conclusion that the control system

resorting to this algorithm is of perfect even in the environment of

serious disturbance.

In this paper, this control law is applied to horizontal and vertical

closed-loop control of AUV CR-02. We made some computer simulations,

and compared with traditional PID controller and sliding controller,

the performance has obviously been improved on accuracy of dynamic

positioning and target tracking, especially under serious disturbance..

Based on simulation, we will carry out a pool test of AUV CR-02.



Authors:H.H. Chen, Z.W. Xing ,Y.Zhang, X.S. Feng