Research on Assembly Method of Threaded Fasteners Based on Visual and Force Information
Round 1
Reviewer 1 Report
The paper presents a work on assembly method of threaded fasteners based on visual and force information. The topic is interesting and the manuscript is well-structured. I suggest publication after revising the text in accordance with the comments given below.
· Please check the text for language and typos.
· Please clearly explain the novelty in this study.
· Introduction should include more discussion about the current technology in the field.
· Fig.2, 5, 6, 7 and 8 are blurry, needs better resolution.
· Are there any other methods to identify these mistakes?
· Is this technique applicable to other processes in manufacturing?
· Is it possible to reduce the detection time?
Minor check is needed for language.
Author Response
Dear reviewer and editor
hello!
First of all, thank you very much for taking the time out of your busy schedule to read and modify my article. Thank you for your valuable suggestions. You have corrected all aspects of the structure, content, research methods and results of my paper which will play a very important role in improving the quality of my paper.
I have carefully studied the comments of the reviewers and carefully revised the paper item by item according to the suggestions.
Our responses to several comments are listed below:
Response Q1: we double-check the language and spelling errors in this article.
Response Q2: The novelty of this study is introduced in detail in line 31 of page 3 ,“The novelty of this paper is to develop an efficient skill learning approach for precision assembly considering solving the above-mentioned problems.……”。
Response Q3: Discussion about the current technology in the field has focused on section II.
Response Q4: These images have been re-edited to improve their resolution.
Response Q5: Other methods for identifying these mistakes in ellipse detection algorithms based on arc segments are described at the beginning of Section 4.
Response Q6: In the revised manuscript, page 19, line 32, we add “In addition, the research of this paper is also applicable to….” to introduce the applications of this technology in other production processes.
Response Q7: In the revised manuscript, page 8, line 14, we add “In the future, numerical optimization can be used to….” to introduce the method to shorten the detection time.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The proposed work establishes an algorithm for detecting the position and orientation of threaded holes and for tightening bolts. The elliptic arc fitting method and the three-point method are used to estimate the initial position and orientation of the threaded hole, and the impact force. Deformation of passive compliance is monitored to control the radial force between the bolt and the threaded hole while tightening.
It would be helpful to provide more details about the vision system and its integration, This would enhance the reproducibility of the proposed algorithm.
Improve the quality of Figures 2, and 4, text is not so clear.
Explain the role of component thickness/depth of the threaded hole.
Include all the list of assumptions followed in this framework.
Although the results are good, it is better to add a comparative assessment with existing literature to draw the merits of the proposed work.
Explain the implication of this method, when different threading sequences are to be analyzed. In addition to the above the following are recommended for further consideration.
Jia, Zhenzhong, et al. "A survey of automated threaded fastening." IEEE Transactions on Automation Science and Engineering 16.1 (2018): 298-310.
Champatiray, Chiranjibi, et al. "Optimal robotic assembly sequence planning with tool integrated assembly interference matrix." AI EDAM 37 (2023): e4.
Author Response
Dear reviewer and editor
hello!
First of all, thank you very much for taking the time out of your busy schedule to read and modify my article. Thank you for your valuable suggestions. You have corrected all aspects of the structure, content, research methods and results of my paper which will play a very important role in improving the quality of my paper.
I have carefully studied the comments of the reviewers and carefully revised the paper item by item according to the suggestions.
Our responses to several comments are listed below:
Response Q1: These images have been re-edited to improve their resolution.
Response Q2: The main research of this paper is the estimation algorithm of the position and orientation of thread hole and position and orientation adjustment in tightening process. Since the final tightening of bolts requires a large tightening torque, the role of thread hole thickness/depth on parts is ignored in the hypothesis, and the compliance control and position and orientation adjustment in tightening process are mainly studied.
Response Q3: In the revised manuscript, page 4, line 19, we add “The research and experiments in this paper follow the following assumptions….” to include all the list of assumptions followed in this framework.
Response Q4: In line 9 on page 19 of this paper, a comparison is made with reference 26.
Response Q5: The research of this paper is aimed at the threaded fastener through the hole, so it has a fixed fastening sequence. In the revised manuscript, page 4, line 19, we add ” And the research of this paper is focused on the threaded fastener of threaded through holes, …” to introduce the different threading sequences is not the main research of this paper.
Author Response File: Author Response.pdf
Reviewer 3 Report
Detachable connection techniques, as used in industrial practice today, are a very important aspect related to manufacturing technology. Savings resulting from the reduction of joining times, while maintaining mechanical properties, are an important aspect of cost optimisation. In this article, the authors focus on developing a method for identifying position of the threaded hole and monitoring and adjusting the fastening process. Presented optical detection method, together with a combination of classical control methods and adaptive control, represents an interesting scientific problem. However, its industrial relevance seems to be low. Detachable joints of machines and equipment are currently designed so that they can be made using present methods that are well established, low-cost and reliable. The problem posed by the authors can also be solved in part by traditional mechanical methods, taking into account the appropriate flexibility (reduction of rigidity in selected axes) of the fastener. Nevertheless, the accuracy and speed of the presented solution should be highly appreciated. In my opinion, the article can be published as it stands.
Author Response
Dear reviewer and editor
hello!
Thank you very much for taking the time out of your busy schedule to read and modify my article. Thank you for your valuable suggestions.
Reviewer 4 Report
Automatic assembly can be significantly facilitated when at least the basics of "Design for Assembly" (DFA) are followed in the design of the assembly.
According to the pictures (Fig. 1)), it is clear that these principles were ignored, and relatively simple actions, which a minor modification of the design could solve, were therefore solved by a complex combination of hardware and software.
The article mentions the image from the camera several times, but as a picture, it does not say what the camera sees.
Author Response
Dear reviewer and editor
hello!
First of all, thank you very much for taking the time out of your busy schedule to read and modify my article. Thank you for your valuable suggestions. You have corrected all aspects of the structure, content, research methods and results of my paper which will play a very important role in improving the quality of my paper.
I have carefully studied the comments of the reviewers and carefully revised the paper item by item according to the suggestions.
Our responses to several comments are listed below:
Response Q1: The purpose of this study is to develop a general assembly method of bolts and threaded holes for different fasteners and assembly scenarios, without the need to modify threaded fasteners. And threaded holes on bolts and threaded fasteners are almost all standard parts, which cannot be modified in design.
Response Q2: In this paper, the images from the camera are mainly used to collect the images of the threaded hole, obtain the center coordinates of the threaded hole and the fitting ellipse of the threaded hole, and finally identify its position and attitude. Section 4.2 of this paper introduces the threaded hole image captured by the camera in Figure 3.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Acceptable.
Reviewer 4 Report
However, I clearly understood what the authors wanted to do, but my opinion is that using the basic methods of DFA is the better and cheaper way in robotics assembly.
The methods that authors do is hardly overkill for these "simple" assembly tasks.
