The underwater glider,by virtue of its autonomy,long life and carrying a variety of sensing devices,can be widely used in marine sampling,submarine topography,ocean platform aided operations and other fields.Compared to a single powerful underwater glider,the relatively simple structure of the multi-underwater-gliders system can improve the efficiency of the task to complete,but also has a higher fault tolerance,and play an increasingly important role in large-scale monitoring and exploration mission.This paper conducts a study of the multi-underwater-gliders system from the following aspects:1.Considering the multi-underwater-gliders system as a virtual multi-body system,the individual is considered as a rigid body in the system.The multi-body system theory is used to analyze the kinematics of the multi-glider system,and in generalized coordinates the Kane method is used to realize the system dynamics modeling.2.Combining the artificial potential field method with the virtual pilot method,the construction method of the multi-gliders system formation is discussed,and a method of constructing a stable glider formation with arbitrary shape is obtained.On this basis,using the rigid body structure formed by the virtual navigator,the glider converges to the desired position under the guidance of the potential field,thus realizing the flexible transformation of multi-glider system formation.3.Based on the discretization of the potential field,the path planning of multi-underwater glider system in obstacle environment is studied.The introduction of the reinforcement learning method makes it possible to solve the problem of path optimization of the multi-gliders system from different locations to a target area at once.And the effectiveness of the method is proved through the simulation experiments. |