Research On Motion Planning Algorithm For Space Robot System With Multi-degree Of Freedom

Robot motion planning is an important problem in robot science and artificial intelligence. Its task is giving a certain task, the appropriate path(s) should be worked out. In recent years, the research of the space robot becomes in the limelight in many countries. And space robots take on more and more space tasks.With the increasing of the robot structure's complexity, some traditional algorithms have presented their shortcomings. This paper addresses the problem of multi-arm space robot motion planning in different environments And Algorithms that is adapt to the multi-degree of freedom robot motion planning should be brought forward.Space robot motion planning is studied in two types of workspace.First, in the workspace which contains obstacles only, Visibility-PRM algorithm has the shortcomings such as low connectivity, low effectiveness in visibility judgment etc. In this paper, we propose an improved Visibility-PRM algorithm (Obs-Visi-PRM) based on reformative obstacle boundary sampling method. This method adopts the strategy of improved sampling and connectivity, thus it can increase the connectivity of roadmap and reduce the collision detection times.Second, aim at the characteristic of the workspace which contains dangerzones, we propose a hierarchical motion planning algorithm. Visibility-PRM is adopted in the first hierarchy; Obs-Visi-PRM is adopted in the second hierarchy; certain improved PRM algorithm can be adopted in the third hierarchy. This method adopts effective sub algorithms in each hierarchy. The improved algorithm is adapt to motion planning problem in workspaces of different complexity and has the advantages like low redundancy and few collision detection times.Some instances are analyzed to verify the new algorithms' effectiveness. And the direction to the research in the future is pointed out.