Research on Path Planning Algorithm of Workpiece Maintenance Manipulator Based on Improved Informed-RRT*

Complex scenarios of workpiece internal and external wall parts installation have strict requirements on the end pose of the maintenance manipulator. Cartesian space path planning is difficult to meet the pose requirements of the end of the manipulator, and it is hard to get inside complex workpieces using the traditional Informed-RRT* algorithm in joint space. Given above problems, this paper proposes an improved Informed-RRT* algorithm based on the node selection strategy after establishing a six-degree-of-freedom manipulator model and performing collision detection analysis. First, a bias strategy based on the neighborhood range of intermediate points and target points is adopted to strengthen the purpose and direction of RRT expansion. In addition, the grandfather node is introduced to turn the curve into a straight line, so that the random tree can converge to the optimal solution faster. Secondly, the node rejection strategy is introduced to reduce the excessive exploration of invalid areas and eliminate useless tree nodes when updating the path, reducing the number of visits to tree nodes. Finally, the correctness and effectiveness of the improved algorithm are verified by simulation experiments. Final simulation results show that the manipulator can enter the specified position inside the workpiece during joint space path planning while solving the shortcomings of the traditional Informed-RRT* algorithm such as blindness, long-running time, and slow convergence speed.