This paper presents a design solution addressing high sampling randomness, slow convergence, and non-smooth paths in traditional RRT algorithms for robotic arm path planning. The proposed approach develops an optimized node sampling strategy to mitigate these issues. Meanwhile, after the algorithm obtains the initial path, trajectory optimization processing is carried out on the path to ensure that it conforms to the dynamic characteristics of the robotic arm. Key enhancements concentrate on strengthening the target orientation of sampling nodes and minimizing invalid spatial exploration, which collectively elevate path planning efficiency. Validation and evaluation will be conducted through the MATLAB simulation platform combined with a six-degree-of-freedom robotic arm physical platform.