Motion Planning For Unmanned Surface Vehicle In Complex Environments

With the rise of the field of artificial intelligence,the research and development of intelligent ship has become the industry consensus.The intelligent and autonomous Unmanned Surface Vehicle（USV）is undoubtedly a research hotspot in recent years.In order to build the intelligent and autonomous system of the USV,the motion planning technology of the USV is one of the key technologies.Especially in the background of the complexity and diversification of the tasks performed by the USV,how to realize the safe and efficient motion planning of the USV is the core problem to be solved.In this thesis,aiming at the motion planning problem of the USV in the complex environments,the research work is carried out by using the research ideas of motion model construction-trajectory unit library construction-motion planning algorithm design-experimental verification,focusing on the motion planning tasks in the multi obstacle environments,restricted water environments,wind and current interference environments,etc.The main research work of this thesis is summarized as follows:First of all,according to the navigation characteristics of the USV,the kinematics and dynamics characteristics of the USV is analysed,and the mathematical model of the ship motion with 3 degrees of freedom as the mathematical model of the USV motion is selected in this study.Secondly,in the multi obstacle environments,the motion planning of the USV is easy to fall into the problem of local optimization and long search time,and a motion planning method based on topological position relationship is proposed.The specific methods are as follows: firstly,analyze the topological position relationship of space objects（such as obstacles,the USV,etc.）;then,analyze the topological position relationship between obstacles and the USV,and build a global topological navigation map;finally,with the mathematical model of USV motion as the constraint,search the optimal path and realize the motion planning of USV facing multi obstacle environments.Thirdly,aiming at the problem that the path planning method considering the kinematic and dynamics characteristics of the ship can not get the optimal solution or not have a solution in the restricted waters environments（complex islands and reefs,narrow and long waters environments）,a motion planning method based on Lable A*algorithm（LAA）is proposed.The specific methods are as follows: firstly,establish the planning space needed for the motion planning of the USV;secondly,establish the LAA search algorithm by improving the traditional A* algorithm;finally,establish the USV rectangle trajectory unit library,combine the rectangle trajectory unit and LAA algorithm to realize the safe and efficient motion planning of the USV in the restricted water environments.Finally,according to the characteristics that the navigation environment of USV is easy to be disturbed by wind and current,a motion planning method of USV considering the influence of wind and current is proposed.The specific methods are as follows: firstly,establish the mathematical model of the motion of the USV considering the wind and current effects,and analyze the influence of wind and current;secondly,update the time-varying trajectory unit library under the influence of the wind and current;thirdly,according to the new trajectory unit library,combined with the requirements of the navigation of the USV,the method extracts the characteristics of the rudder angle and the navigation direction angle of the trajectroy to constrain the path search;finally,According to the influence of wind and current,time-varying trajectory unit and trajectory characteristics,the motion planning method of USV is established to realize the motion planning of USV based on the time-varying trajectory unit.Experiments show that the path of motion planning method considering wind and current effects is more close to the actual trajectory of USV in complex environments.In this thesis,the motion planning of USV in multiple application scenarios is discussed,and according to the different navigation environments of the USV,the corresponding solutions are put forward to help the USV to achieve autonomous navigation in the complex environments,which can further increase assistance for the intelligent navigation and even the intelligent ship,and has certain positive significance for the realization of the intelligent navigation of the ship.