Obtaining the position and posture information of regular stacking stick is the key to guide the manipulator to grab. The obtained 3D point cloud of stacking stick is processed through filtering to obtain the point cloud of top sticks

and it is segmented to get the radius

position and posture of each stick after filtered with the random sample consensus（ RANSAC） algorithm. The simulation results show the correctness of the recognition results of the method. By scanning the regular stacking stick with the diameter of 35. 00 mm and the length of80. 00 mm

the point cloud of the same stick is segmented multiple with different parameters. The experimental results show that when the number of iterations k is 20 and the distance threshold t is 0. 7

the segmentation result is stable and consumes about 160 ms. The maximum error of stick radius is 0. 47 mm. This method has some robustness and feasibility. The method can accurately recognize the position and posture of regular stacking stick

can be widely used in the position and posture recognition of rod parts

and then guide the manipulator grabbing.