Abstract: An improved rapidly-expanding random tree (RRT) algorithm was proposed to plan the path of a six degree of freedom (DOF) robotic arm in response to the problem of high path cost and slow sampling process in the RRT algorithm. Combining the advantages of RRT * and RRT-connect algorithms, the target sampling was applied to enhance the tendency of the algorithm to search for target points, the greedy thinking was introduced to improve the efficiency of the algorithm, and the quintic B-spline interpolation was combined to optimize the path smoothness, shorten the path planning time and length. The matplotlib function pack in Python was used to calculate the time, number of sampling points, and path length required for RRT and improved algorithm planning, as well as the dynamic planning library for algorithm configuration on the ROS platform was used to verify the path planning effect of the improved algorithm. The results show that compared with the traditional RRT algorithm, the path planning of the six DOF robotic arm by the proposed improved algorithm shortens the average path length by 29.5%, the planning path time is shortened by 8.5%, and the success rate of path planning is increased to 96.7%,which verifies the feasibility of the algorithm in practice.