Research On 6-Dof Robot Seam Tracking Based On Structured Light Vision

In the modern welding manufacturing, it is difficult to meet the production requirements on high quality and efficiency with teaching-playback welding robots, and it is a certain trend to achieve robot welding automation. At present, the method based on structured light sensors has become the most effective way of seam tracking and has attracted great attention for its advantages of high measurement precision and anti-jamming. In this paper, the 6-dof robot seam tracking system based on structured light vision is studied and the main contents of this paper are included as follows. 1) Research on calibration of the robot seam tracking systemThe visual system based on structured light is studied and the light plane is calibrated by using direct calibration method. In view of the complexities of the calibration between multiple systems, it is required to analyze the relationship between each subsystem, and calibrate the robot seam tracking system by using the calibrated vision system and robot system as measurement tools. 2) Algorithms research on detection of feature points of seam imagesThe extraction algorithms of light stripe center lines of seam images are compared and analyzed. The image processing algorithms based on two types of seams are proposed in this paper, including image preprocessing, extraction of center lines through a rapid and accurate method, identification of feature points through detection of endpoints and intersection points. Finally, the real-time performance and reliability of the image processing system with uncertainty disturbance are tested by some experiments. 3) Research on seam tracking methodA seam tracking method based on dog-leg seam is proposed, which can be operated by three steps, including pre-measurement of the seam profile, identification of inflection points and determination of corner parts as well as seam tracking in real time. In order to guarantee the accuracy and stability of the system in the case of external disturbance, the automatic correction is adopted to decrease the effect due to assembly error in real production. 4) Robot seam tracking experiments and analysisThe experimental platform is designed and built in this paper. Several experiments are performed to verify the accuracy of the calibration of the system, the reliability of the visual system and the validity of the automatic correction. Moreover, the traditional automatic tracking method and proposed method are adopted respectively in the robot seam tracking experiments in different speed to test the performance of the robot seam tracking system.