Picking Robot Visual Servo Control Based on Modified Fuzzy Neural Network Sliding Mode Algorithms
Round 1
Reviewer 1 Report
The authors propose a joint visual servo control algorithm based on improved sliding mode, kinematics & dynamics equation. Overall the paper is well written with appropriate discussion. Reviewer has few comments to improve the manuscript,
"However, sliding mode 15 control can bring jitter and other problems."What are the other major problems. Please include them.
Few more recent related works needs to be discussed in the Introduction section. Highlight their limitations and issues to strengthen the novelty of your work.
Discuss the novelty of the work as a separate paragraph in the Introduction part and highlight the contribution of your work as separate bullet points.
Include organization of the paper towards the end of Introduction section.
Clarity of figure 1 and figure 2 to be improved
Font size in figures 5,6,7,8,9,10,11,12,13 to be increased for better readability.
The paper has few grammatical errors and Typos that needs to be eliminated. Reviewer suggests an English Proofreading.
The proposed system, analysis and discussion is convincing. Reviewer doesn't have major comments on this.
Clarity of the graphs to be improved.
What are the limitations of the proposed approach?Discuss
Future works is missing.
Author Response
Response to Reviewer 1 Comments
Dear Editors and Reviewer:
Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled
“Picking Robot Visual Servo Control Based on Modified Fuzzy Neural Network Sliding Mode Algorithms” (ID: electronics-501109). Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied these comments carefully and have made corrections which we hope meet with approval. The main corrections in the paper and the responds to the reviewer’s comments are as flowing:
Responds to the reviewer’s comments:
Point 1: "However, sliding mode 15 control can bring jitter and other problems." What are the other major problems. Please include them.
Response 1: Thank you for your instructive suggestions. The biggest problem of sliding mode variable structure control is chattering, and the speed, inertia, acceleration, switching surface and other factors are also considered when approaching the sliding die surface. I have made changes in the article.
Point 2: Few more recent related works needs to be discussed in the Introduction section. Highlight their limitations and issues to strengthen the novelty of your work.
Response 2: Thank you for your proposal. I have added some current research status and analyzed their strengths and weaknesses in the introduction.
A great deal of research has been devoted to these intelligent algorithm scholars, but there are few applications in the field of agricultural picking robots. In this paper, a joint visual servo control algorithm based on improved sliding mode and kinematics & dynamics equation is presented which appears to a well robustness in non-linear system. However, sliding mode control always requires detailed system information and corresponding uncertainty bounds to ensure stability[18]. Even though the stability requirements are met, there will still be problems such as jitter. So the fuzzy and adaptive neural network control algorithms are introduced, since the traditional neural network model is too complex with a large amount of calculation, a Dropout layer is added to hide part of the data randomly so that overfitting can be avoided. And combined with sliding mode control to suppress the jitter.
According to the characteristics of apple picking robot motion control, the image-based robot visual servo control method is combined with the improved fuzzy neural network sliding mode control algorithm to make use of the nonlinear and non-system-dependent mechanism model of fuzzy neural network. Combined with sliding mode variable structure control, the sliding mode chattering is eliminated, and the stability of the control system is improved. In this paper, the algorithm is applied for the first time in the field of agricultural fruit picking robot, which improves the extraction speed of apple picking robot.
Point 3: Discuss the novelty of the work as a separate paragraph in the Introduction part and highlight the contribution of your work as separate bullet points.
Response 3: Thank you for your proposal. I have already added an analysis of the novelty of the work in the introduction, highlighting the superiority of this article.
According to the characteristics of apple picking robot motion control, the image-based robot visual servo control method is combined with the improved fuzzy neural network sliding mode control algorithm to make use of the nonlinear and non-system-dependent mechanism model of fuzzy neural network. Combined with sliding mode variable structure control, the sliding mode chattering is eliminated, and the stability of the control system is improved. In this paper, the algorithm is applied for the first time in the field of agricultural fruit picking robot, which improves the extraction speed of apple picking robot.
Point 4: Include organization of the paper towards the end of Introduction section.
Response 4: I have added a new paragraph to the introduction to the paper structure.
The research of this paper is mainly to verify the visual servo control of the sliding mode control algorithm based on improved fuzzy neural network through apple picking robot. The article is divided into three parts. The article is divided into three parts, Firstly, establishing the kinematics and dynamics equations of the picking robot. Secondly, visual positioning is introduced to calculate the position of the target point, and an image-based visual control algorithm is used. Finally, the sliding mode control algorithm and the improved fuzzy neural network control algorithm are combined to carry out simulation analysis, and the algorithm is verified on the picking robot. It is concluded that the improved fuzzy neural network sliding mode control algorithm can improve the efficiency of the robot arm servo control and has higher stability.
Point 5: Clarity of figure 1 and figure 2 to be improved
Response 5: I have made modified in the article.
Point 6: Font size in figures 5,6,7,8,9,10,11,12,13 to be increased for better readability.
Response 6: Thank you for your proposal. I have made the corresponding modified in the article.
Point 7: The paper has few grammatical errors and Typos that needs to be eliminated. Reviewer suggests an English Proofreading.
Response 7: Thank you!
Point 8: The proposed system, analysis and discussion is convincing. Reviewer doesn't have major comments on this.
Response 8: Thank you!
Point 9: Clarity of the graphs to be improved.
Response 9: I have made the corresponding modified in the article.
Point 10: What are the limitations of the proposed approach? Discuss.
Response 10: The shortcoming of this paper is the algorithm verification work carried out in the laboratory environment, and the related test work is not carried out in the real picking environment. Due to the low stability of the mechanical design of the current picking robot end effector, we will improve the existing mechanical structure of the picking robot in the future work, and continue to verify the superiority of the current algorithm on the basis of the new terminal actuator, and complete the picking experiment in the real environment of the apple picking robot.
Point 11: Future works is missing.
Response 11:
I added the work to be done in the future in the conclusion.
Due to the low stability of the mechanical design of the current picking robot end effector, we will improve the existing mechanical structure of the picking robot in the future work, and continue to verify the superiority of the current algorithm on the basis of the new terminal actuator, and complete the picking experiment in the real environment of the apple picking robot.
Reviewer 2 Report
The article deals with an attempt to improve servo control algorithm of a fruit picking robot.
Agriculture robotics is indeed a very significant field of study, so papers in this area are interesting and important.
I have several comments that must be addressed before publishing this paper:
Claims in lines 32 and 54-59 must be referenced by relevant literature.
Lines 32-36: unclear sentence
Lines 39-40: unclear sentence
Is fig. 1 was taken from somewhere or is it your original drawing?
The manipulator/griper was mentioned several times but there is no evidence of it in the schematic drawings. Authors must include the schematic geometry of this griper at the beginning of the paper.
Add some pictures of the real robot and also expend the description of it at the beginning of the paper. Was this robot developed by you? If not, which robot is it?
Don't let the reader wait till the end of the paper to see that there are actual pictures of the real robot rather than just schematic drawings.
The results and conclusions are not sufficient, authors must include quantitative results in these sections. How much the performance was improved?
Author Response
Major corrections and reviewers'comments in the paper are attached.
Author Response File: 
Author Response.pdf
Reviewer 3 Report
Interesting work. It presents relevant concepts from the literature and real approaches that complement the results of computational simulation. The following are some aspects for improving the quality of the paper.
Abstract.
It presents relevant characteristics of what will be treated and the contextualization of the problem.
He failed to present values obtained in the tests to confirm the statements presented in the abstract.
1 Introduction.
The introduction presents essential concepts about the evaluated context, but I have lacked related work of fuzzy neural network for robot controls. This would leave the reader more confident that the approach is consistent for solving problems. I indicate as models that meet the requirement:
@article{kiguchi2000position,
title={Position/force control of robot manipulators for geometrically unknown objects using fuzzy neural networks},
author={Kiguchi, Kazuo and Fukuda, Toshio},
journal={IEEE Transactions on Industrial Electronics},
volume={47},
number={3},
pages={641--649},
year={2000},
publisher={IEEE}
}
@article{gao2001adaptive,
title={Adaptive control of robot manipulators using fuzzy neural networks},
author={Gao, Yang and Er, Meng Joo and Yang, Song},
journal={IEEE Transactions on Industrial Electronics},
volume={48},
number={6},
pages={1274--1278},
year={2001},
publisher={IEEE}
}
@article{er2003robust,
title={Robust adaptive control of robot manipulators using generalized fuzzy neural networks},
author={Er, Meng Joo and Gao, Yang},
journal={IEEE Transactions on Industrial Electronics},
volume={50},
number={3},
pages={620--628},
year={2003},
publisher={IEEE}
}
@article{he2018adaptive,
title={Adaptive fuzzy neural network control for a constrained robot using impedance learning},
author={He, Wei and Dong, Yiting},
journal={IEEE transactions on neural networks and learning systems},
volume={29},
number={4},
pages={1174--1186},
year={2018},
publisher={IEEE}
}
@article{camci2018aerial,
title={An aerial robot for rice farm quality inspection with type-2 fuzzy neural networks tuned by particle swarm optimization-sliding mode control hybrid algorithm},
author={Camci, Efe and Kripalani, Devesh Raju and Ma, Linlu and Kayacan, Erdal and Khanesar, Mojtaba Ahmadieh},
journal={Swarm and evolutionary computation},
volume={41},
pages={1--8},
year={2018},
publisher={Elsevier}
}
Missed the final paragraph explaining the organization of the paper for the reader.
2 Picking Robot Structure and Model Establishment
It presents relevant aspects of the formatting and construction of the robot used in the paper. There is a lack of scientific standardization, such as the numbering of all equations present in the text.
3. Visual Servo Control
No comments.
4. Complex Fuzzy Neural Network Sliding Mode Control
There is a lack of concepts and explanations about fuzzy neural networks. A better explanation of these elements is essential for readers of the journal, especially with some examples and applications.
I advise some tips on the fuzzy neural network:
Examples of current fuzzy neural network models:
@article{de2019data,
title={Data density-based clustering for regularized fuzzy neural networks based on nullneurons and robust activation function},
author={de Campos Souza, Paulo Vitor and Torres, Luiz Carlos Bambirra and Guimaraes, Augusto Junio and Araujo, Vanessa Souza and Araujo, Vincius Jonathan Silva and Rezende, Thiago Silva},
journal={Soft Computing},
pages={1--15},
publisher={Springer}
}
Recent articles that fuzzy neural networks act on health problems.
@article{junio2019pruning,
title={Pruning Fuzzy Neural Network Applied to the Construction of Expert Systems to Aid in the Diagnosis of the Treatment of Cryotherapy and Immunotherapy},
author={Junio Guimar{\~a}es, Augusto and Vitor de Campos Souza, Paulo and Jonathan Silva Ara{\'u}jo, Vin{\'\i}cius and Silva Rezende, Thiago and Souza Ara{\'u}jo, Vanessa},
journal={Big Data and Cognitive Computing},
volume={3},
number={2},
pages={22},
year={2019},
publisher={Multidisciplinary Digital Publishing Institute}
}
@article{silva2019using,
title={Using Resistin, Glucose, Age and BMI and Pruning Fuzzy Neural Network for the Construction of Expert Systems in the Prediction of Breast Cancer},
author={Silva Ara{\'u}jo, Vin{\'\i}cius Jonathan and Guimar{\~a}es, Augusto Junio and de Campos Souza, Paulo Vitor and Silva Rezende, Thiago and Souza Ara{\'u}jo, Vanessa},
journal={Machine Learning and Knowledge Extraction},
volume={1},
number={1},
pages={466--482},
year={2019},
publisher={Multidisciplinary Digital Publishing Institute}
}
Models with other approaches of fuzzy neural networks:
@article{hsieh2019single,
title={Single index fuzzy neural networks using locally weighted polynomial regression},
author={Hsieh, Jer-Guang and Jeng, Jyh-Horng and Lin, Yih-Lon and Kuo, Ying-Sheng},
journal={Fuzzy Sets and Systems},
year={2019},
publisher={Elsevier}
}
@article{gaxiola2019pso,
title={PSO with Dynamic Adaptation of Parameters for Optimization in Neural Networks with Interval Type-2 Fuzzy Numbers Weights},
author={Gaxiola, Fernando and Melin, Patricia and Valdez, Fevrier and Castro, Juan R and Manzo-Mart{\'\i}nez, Alain},
journal={Axioms},
volume={8},
number={1},
pages={14},
year={2019},
publisher={Multidisciplinary Digital Publishing Institute}
}
@article{xu2019synchronization,
title={Synchronization Control Algorithm of Double-Cylinder Forging Hydraulic Press Based on Fuzzy Neural Network},
author={Xu, Xiaodan and Bai, Zhifeng and Shao, Yuanyuan},
journal={Algorithms},
volume={12},
number={3},
pages={63},
year={2019},
publisher={Multidisciplinary Digital Publishing Institute}
}
@article{skowron2019application,
title={Application of Self-Organizing Neural Networks to Electrical Fault Classification in Induction Motors},
author={Skowron, Maciej and Wolkiewicz, Marcin and Orlowska-Kowalska, Teresa and Kowalski, Czeslaw T},
journal={Applied Sciences},
volume={9},
number={4},
pages={616},
year={2019},
publisher={Multidisciplinary Digital Publishing Institute}
}
Is the model presented itself or is it based on another approach? If so, what are the reasons for the model choices? Why does it stand out from other approaches?
Conclusions
There was a lack of future work to be done on this excellent approach.
I would like to see in the conclusions greater emphasis on the gains of the utilization of the fuzzy neural network.
General Tips:
The equations must be numbered correctly, starting from topic 2.2.
Improve the quality of figures 8, 9, 10, 11, 12 and 13.
Author Response
Response to Reviewer 3 Comments
Dear Editors and Reviewer:
Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled
“Picking Robot Visual Servo Control Based on Modified Fuzzy Neural Network Sliding Mode Algorithms” (ID: electronics-501109). Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied these comments carefully and have made corrections which we hope meet with approval. The main corrections in the paper and the responds to the reviewer’s comments are as flowing:
Responds to the reviewer’s comments:
Point 1: He failed to present values obtained in the tests to confirm the statements presented in the abstract.
Response 1: Thank you for your instructive suggestions. We enriched the laboratory data, made corresponding comparison experiments, and did experimental analysis, highlighting the superiority of the algorithm.
Point 2: The introduction presents essential concepts about the evaluated context, but I have lacked related work of fuzzy neural network for robot controls. This would leave the reader more confident that the approach is consistent for solving problems. I indicate as models that meet the requirement.
Response 2: Thank you for your proposal. I have added some current research status and analyzed their strengths and weaknesses. The corresponding modifications have been made in the paper.
Point 3: Missed the final paragraph explaining the organization of the paper for the reader.
Response 3: I have added a new paragraph to the introduction to the paper structure.
The research of this paper is mainly to verify the visual servo control of the sliding mode control algorithm based on improved fuzzy neural network through apple picking robot. The article is divided into three parts. The article is divided into three parts, Firstly, establishing the kinematics and dynamics equations of the picking robot. Secondly, visual positioning is introduced to calculate the position of the target point, and an image-based visual control algorithm is used. Finally, the sliding mode control algorithm and the improved fuzzy neural network control algorithm are combined to carry out simulation analysis, and the algorithm is verified on the picking robot. It is concluded that the improved fuzzy neural network sliding mode control algorithm can improve the efficiency of the robot arm servo control and has higher stability.
Point 4: There is a lack of scientific standardization, such as the numbering of all equations present in the text.
Response 4: This article has been modified.
Point 5: There is a lack of concepts and explanations about fuzzy neural networks.
Response 5: I have explained the concept of fuzzy neural networks in lines 230-235.
Fuzzy neural network is a kind of special neural network which is a hybrid intelligent system formed by the combination of neural network and fuzzy logic. It combines the two kinds of techniques by combining the human-like reasoning of fuzzy systems with the learning and connection structure of neural networks. In a nutshell, the fuzzy neural network (FNN) assigns a conventional neural network to fuzzy input signals and fuzzy weights. Its function is to use fuzzy neural network structure to implement fuzzy logic reasoning.
Point 6: Is the model presented itself or is it based on another approach? If so, what are the reasons for the model choices? Why does it stand out from other approaches?
Response 6: The model is presented by itself. According to the characteristics of apple picking robot motion control, the image-based robot visual servo control method is combined with the improved fuzzy neural network sliding mode control algorithm to make use of the nonlinear and non-system-dependent mechanism model of fuzzy neural network. Combined with sliding mode variable structure control, the sliding mode chattering is eliminated, and the stability of the control system is improved.
Point 7: There was a lack of future work to be done on this excellent approach.
Response 7: I added the work to be done in the future in the conclusion.
Due to the low stability of the mechanical design of the current picking robot end effector, we will improve the existing mechanical structure of the picking robot in the future work, and continue to verify the superiority of the current algorithm on the basis of the new terminal actuator, and complete the picking experiment in the real environment of the apple picking robot.
Point 8: I would like to see in the conclusions greater emphasis on the gains of the utilization of the fuzzy neural network.
Response 8: We have continued to follow up on the project work, followed by several trials and comparisons of different methods, and the article supplemented the detailed relevant test results. The superiority of the proposed algorithm is obtained through experiments.
We have carried out a lot of experiments on the apple picking robot, using the conventional PID algorithm, the sliding mode control algorithm and the improved fuzzy neural network sliding mode control on the robot arm servo control, in the same experimental environment. Comparing the above three different algorithms on the picking robot system on the grab speed and the grab accuracy, Table 3 is the data comparison using different algorithms.
Table 3. Experimental result
Algorithm | Success rate | Starting position to the end of the crawl time（s） |
Conventional PID algorithm | 88.2% | 15.8 |
Sliding mode control algorithm | 74.0% | 13.0 |
improved fuzzy neural network sliding mode control algorithm | 91.2% | 13.8 |
Through the comparison of experimental data, it can be known that the conventional PID algorithm captures a long time, while the sliding mode control algorithm can effectively reduce the grab time, but there is a lower catch success rate, and the improved fuzzy neural network algorithm is adopted. The sliding mode control algorithm not only improves the grabbing efficiency, but also greatly improves the success rate of the capture. It is shown in the result of visual servo motion of the manipulator that the fruit picking robot based on improved fuzzy neural network sliding mode algorithm has good stability and robustness.
Point 9: The equations must be numbered correctly, starting from topic 2.2.
Response 9: I have made the corresponding modified in the article.
Point 10: Improve the quality of figures 8, 9, 10, 11, 12 and 13.
Response 10: Thank you for your proposal. I have made modified in the article.
Round 2
Reviewer 2 Report
The paper has been improved and the majority of my comments have been addressed. I recommend accepting it after a minor revision.
Consider these minor comments:
Still some claims must be properly referenced in page 2:
1. "Even though the stability requirements are met, there will still be problems such as jitter."
2. "So the fuzzy and adaptive neural network control algorithms are introduced, since the traditional neural network model is too complex with a large amount of calculation, a Dropout layer is added to hide part of the data randomly so that overfitting can be avoided." - in addition to referencing it, this sentence should be also rephrased.
and in page 1:
"The picking robot manipulator is a typical non-linear system because it is composed of multiple joints and has time-varying uncertainty in the picking process so that traditional control method may not meet the control requirements." - This sentence must also be rewritten and grammatically checked, since very confusing.
Another grammar error that must be corrected:
"And combined with sliding mode control to 75 suppress the jitter." - new sentence cannot start with "and".
Overall, the writing style should be consistent. For example, in the conclusion section, the sentence "we will improve the existing mechanical structure of the picking robot in the future work.." should be "the existing mechanical structure of the picking robot will be improved in future work". The same comment is for "We have carried out a lot of experiments on the apple picking robot..." in page 16.
Author Response
Response to Reviewer 2 Comments
Dear Editors and Reviewer:
Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled
“Picking Robot Visual Servo Control Based on Modified Fuzzy Neural Network Sliding Mode Algorithms” (ID: electronics-501109). We have studied these comments carefully and have made corrections which we hope meet with approval. The main corrections in the paper and the responds to the reviewer’s comments are as flowing:
Responds to the reviewer’s comments:
Point 1:
1. "Even though the stability requirements are met, there will still be problems such as jitter."
2. "So the fuzzy and adaptive neural network control algorithms are introduced, since the traditional neural network model is too complex with a large amount of calculation, a Dropout layer is added to hide part of the data randomly so that overfitting can be avoided." - in addition to referencing it, this sentence should be also rephrased.
Response 1: Thank you for your instructive suggestions. I have added references to the sections described above. And the second part of the sentence has been rephrased, as follows:
Therefore, fuzzy and adaptive neural network control algorithms are introduced and combined with sliding mode control to suppress the sliding mode jitter problem.
However, traditional neural network model is complex and computationally intensive, with high probability of over-fitting. To prevent over-fitting , we increase the dropout rate for the Dropout layer.
Point 2: "The picking robot manipulator is a typical non-linear system because it is composed of multiple joints and has time-varying uncertainty in the picking process so that traditional control method may not meet the control requirements." - This sentence must also be rewritten and grammatically checked, since very confusing.
Response 2: Thank you for your proposal. I have made modified in the article, as follows:
The picking robot manipulator is a typical non-linear system because it is composed of multiple joints and has time-varying uncertainty during the picking process. This indicates that traditional control method would not meet the control requirements.
Point 3: "And combined with sliding mode control to 75 suppress the jitter." - new sentence cannot start with "and".
Overall, the writing style should be consistent. For example, in the conclusion section, the sentence "we will improve the existing mechanical structure of the picking robot in the future work.." should be "the existing mechanical structure of the picking robot will be improved in future work". The same comment is for "We have carried out a lot of experiments on the apple picking robot..." in page 16.
Response 3: Thank you for your proposal. This article has been modified.
