A Coarse-to-Fine Method for Estimating the Axis Pose Based on 3D Point Clouds in Robotic Cylindrical Shaft-in-Hole Assembly

Abstract

In this work, we propose a novel coarse-to-fine method for object pose estimation coupled with admittance control to promote robotic shaft-in-hole assembly. Considering that traditional approaches to locate the hole by force sensing are time-consuming, we employ 3D vision to estimate the axis pose of the hole. Thus, robots can locate the target hole in both position and orientation and enable the shaft to move into the hole along the axis orientation. In our method, first, the raw point cloud of a hole is processed to acquire the keypoints. Then, a coarse axis is extracted according to the geometric constraints between the surface normals and axis. Lastly, axis refinement is performed on the coarse axis to achieve higher precision. Practical experiments verified the effectiveness of the axis pose estimation. The assembly strategy composed of axis pose estimation and admittance control was effectively applied to the robotic shaft-in-hole assembly.

Calibration and pose measurement of a combined vision sensor system for industrial robot grasping of brackets

“…However, image-processing algorithms depend on a reliable environment and lack flexibility owing to variable parameters. Li et al [19] proposed a coarse-tofine axis pose estimation method for robot assemblies based on three-dimensional (3D) point clouds. The shafts and holes were reconstructed using 3D cameras.…”

Quality inspection and assembly sequence optimization of revolved thin-walled parts based on point clouds

“…Currently, virtual assembly technology based on three-dimensional point clouds mainly includes two directions: first, combining point cloud data with mechanical models to complete virtual assembly and deformation analysis [1] ; second, directly processing three-dimensional point cloud data, performing operations such as 3D reconstruction, registration, surface, and curve fitting on the point cloud to achieve assembly and assembly quality inspection [2,3,4,5,6,7,8] . While methods combining mechanical model analysis of component deformation can predict errors transmitted during the assembly process, accurate estimates of material mechanical properties and the forces acting on the assembly structure are required.…”

Research of Online Hand–Eye Calibration Method Based on ChArUco Board

“…However, the traditional working methods of industrial robots have been unable to meet the increasing production demand. Industrial robots equipped with vision sensors perceive the surrounding environment through image processing technology, which improves the “flexibility” of robot operations, and enables robots to complete more complex production tasks [ 3 , 4 , 5 ]. Hand–eye calibration is an indispensable part of the visual servo operation of the robot, which is an important bridge between the robot and the vision sensor.…”