Research On The Construction Of Knowledge Barrier Wall Based On Artificial Moment Method In Three Dimensional Environment

Path planning is one of the important research fields of robotics,and it is an important guarantee for mobile robots to complete tasks smoothly.With the continuous research in related fields,the path effect is constantly optimized,but the application environment of most algorithms is set to two-dimensional environment.Compared with the actual three-dimensional environment,the movement of the robot in the two-dimensional environment is relatively simple,the obstacle setting is too ideal,and the complexity is relatively low.Thus,although the theory of two-dimensional path planning algorithm can be applied in three-dimensional environment,the actual application still needs to undergo great changes.At present,the Artificial Moment Method has been widely used in the two-dimensional environment,but it has not been successfully applied in the three-dimensional environment,so this thesis focuses on this problem.The main work of this thesis is as follows:First,based on the theory and related concepts of the Artificial Moment Method,as well as the nature of the robot in the three-dimensional environment,the basic model of the Artificial Moment Method in the three-dimensional environment is built,and the difference between some concepts of the Artificial Moment Method in the two-dimensional environment and the three-dimensional environment is analyzed,so as to control the application of the Artificial Moment Method in the three-dimensional environment on a macro level.Secondly,the construction of knowledge obstacle wall in three-dimensional environment in the Artificial Moment Method is discussed.According to the theorem that the interface area between the sphere and the plane must be circular,the actual recognition area of the robot on the obstacle wall can be completed as a circle,and the completion circle is the interface area between the sphere recognition range of the robot and the plane where the obstacle plane is located.Therefore,the actual recognition region of the robot can also be expressed as the interface region between the completed circle and the obstacle plane,and the three-dimensional problem can be transformed into a two-dimensional problem by using this method.Since the path of the robot in the three-dimensional environment is more complex,the knowledge obstacle wall constructed according to the actual identification area of the robot is also more complex.In this thesis,the boundary points of the identification area are recorded to construct and record the corresponding knowledge obstacle wall.Third,aiming at the concrete construction steps of knowledge obstacle wall,this thesis proposes Hexagon Representation Method for the first time.A proper regular hexagon is constructed according to the completion circle,and the arcs that are not conducive to record are converted through the six boundary line segments of the hexagon.Finally,the construction and recording of the knowledge obstacle wall are completed through the coordination and connection of the recognition area,regular hexagon and obstacle surface.Fourthly,according to the different construction principles,the Hexagon Representation Method is divided into the internal inclusion method and the external expansion method.,and according to the different construction conditions,each method is divided into two cases.The specific construction rules of the four cases of the two methods are expounded and analyzed in combination with the schematic diagram.Finally,simulation experiments are carried out based on the Hexagon Representation Method proposed in this thesis.The feasibility of the method is verified by the simulation of various basic conditions,and the convenience of the method is verified by comparing the experimental data with the raster data.Finally,the construction and recording of knowledge obstacle wall in 3D environment are successfully completed.