What is it about?
The reliability of the transmission system hinges significantly on the assembly quality of its main component, the large gear structures. However, the traditional approach employing manual lifting presents a host of challenges, such as high assembly complexity and lowered efficiency, rendering the overall assembly process notably arduous. In this study, a large gear structure assembly method based on uncalibrated image visual servo guidance is proposed. Comprising three modules, the approach involves constructing a task function for projective homography, estimating the image Jacobian matrix, and designing an adaptive servo controller. This methodology facilitates the mapping of changes in gear images to the motion of the end-effector in the parallel mechanism. Consequently, the system dynamically guides the end-effector to achieve the required attitude adjustments in the gear assembly in response to changes in the image features. Experimental results demonstrate that the method proposed surpasses alternative approaches, simultaneously exhibiting a significant enhancement in assembly efficiency. The method has a wide application prospect in the field of automated assembly of large gear structures.
Featured Image
Photo by Daniel Apodaca on Unsplash
Why is it important?
The main contributions of this paper can be summarised as follows: (1) A improved task function for projective homography is introduced, aiming to significantly minimize the error associated with pixel point matching within the homography matrix, thereby enhancing robustness. (2) A Kalman filter Jacobi matrix estimator with memory attenuation is designed, incorporating an exponential attenuation factor to mitigate external noise interference. This improvement contributes to heightened system accuracy and increased resistance to interference. (3) An adaptive servo controller, employing a Q-learning framework, is advanced to uphold system precision while concurrently enhancing servo efficiency. (4) Experimental validation shows the excellent performance of the proposed method in terms of both efficiency and accuracy. This approach offers a pioneering solution to the intricate challenge of positioning and assembling gears or other rotary components.
Perspectives
In this study, a large gear structure assembly method based on uncalibrated image visual servo guidance is proposed. The primary objective is to overcome the challenges in traditional assembly methods, focusing on the difficulties and inefficiencies experienced in the assembly process. Firstly, the homography estimation method is optimized, and the optimized pixel-point matching error is reduced by about 2/3. Based on this, the task function of image serving is constructed. Subsequently, an exponentially weighted attenuation factor is introduced into the Kalman filter and applied to the online estimation of the image Jacobian matrix. It is able to suppress environmental noise, such as camera and light, and improve the performance of the servo system. In addition, an adaptive servo controller is designed based on Q-learning, which effectively improves the efficiency of the visual servo system. Finally, the excellent performance of the proposed method is demonstrated by two types of experiments. In the calibration board movement experiment, the proposed method obtains a significant improvement in computational efficiency compared to the classical IBVS method. In the gear assembly experiment, the proposed method is much better than the other four SOTA methods about servo accuracy. Moreover, compared with the traditional manual assembly method, the assembly efficiency is improved by five times. This provides a novel solution for the accurate positioning and assembly of gears and rotary parts.
Xiang Huang
Nanjing University of Aeronautics and Astronautics
Read the Original
This page is a summary of: Large gear structure assembly method based on uncalibrated image visual servo guidance, Review of Scientific Instruments, December 2023, American Institute of Physics,
DOI: 10.1063/5.0177035.
You can read the full text:
Contributors
The following have contributed to this page
