Aiming at the problem of how to control an open-frame underwater vehicle precisely,a new method of control based on variable structure neural network was proposed.

When an autonomous underwater vehicle is in marine survey,in order for it to adjust to a wide variety of situations rapidly,replanning decisions must be taken to find the best scheme to finish the mission satisfactorily and ensure the vehicle s safety.

It is necessary for AUV(autonomous underwater vehicle) to have the ability of obstacles-avoidance during a long range travel in the unknown ocean environment.

The ability of obstacle avoidance based on forward looking sonar is a basic requirement of AUV(autonomous underwater vehicle) during a long range travel in the unknown ocean environment.

Basing on the terrain scanning mission,this paper focuses on the problem of mission and task coordination method at the planning layer of AUV(Autonomous Underwater Vehicle) control system.

A new dynamic obstacle avoiding method of autonomous underwater vehicle(AUV) based on the sensing information is proposed.

Aimed at the problems of motion planning for autonomous underwater vehicle(AUV) in unknown environment,a behavior-based approach of obstacle avoidance and goal approaching is raised.

In order to insure that autonomous underwater vehicle completes the tasks successfully in unstructured and hazardous oceanic environment with the high pressure and low visibility, an efficient and effective fault-tolerant control system becomes imperative for AUVs.

The autonomous control system of autonomous underwater vehicle(short for AUV)is very important for safty and validity of mission execution.

SPACE MOTION EQUATIONS FOR SAUV;