Collaborative Robots with Cognitive Capabilities for Industry 4.0 and Beyond
Round 1
Reviewer 1 Report (Previous Reviewer 1)
Comments and Suggestions for AuthorsGeneral remark. The entire research claims direction of evolving of collaborative robots as single path, but reality more complex. I would encourage authors to think about development as a spread process rather than a straight line.
This paper seems to me a review, but not a research paper. This should be noted in the title or journal classification.
Not mentioned of other robot-human relation, such as cooperation mode, coexistence, and synchronization.
Issue of synthetic ontology, not mentioned at least in cognition issue.
Conclusions contains no statement from authors, so the outcome of the entire analysis remain unclear.
Comments on the Quality of English Language
English looks generally OK.
Author Response
Comment 1: General remark. The entire research claims directionof evolving of collaborative robots as single path, but reality more complex. I would encourage authors to think about development as a spread process rather than a straight line.
Response 1 We completely agree that development should be conceived as a spread process rather than a straight line, as suggested by figure 3. However, the figure is a simplification and the dashed straight (but growing) line is simply a kind of regression line of a ragged process. We added a sentence on this purpose in page 9 of the manuscript (highlighted in yellow).
Comment 2: This paper seems to me a review, but not a research paper. This should be noted in the title or journal classification.
Response 2: The paper is classified by the editorial board as Perspective.
Comment 3: Not mentioned of other robot-human relation, such as cooperation mode, coexistence, and synchronization.
Response 3: We agree that these are relevant and interesting issues that become crucial at the implementation level.
Comment 4: Issue of synthetic ontology, not mentioned at least in cognition issue.
Response 4: In the second section (Cognition vs. Intelligence) we address the fundamental issue of epistemology. There is no doubt that in general philosophical terms epistemology and ontology are linked: Ontology is concerned with truth and thus the representation of knowledge; Epistemology is focused on knowledge acquisition by a cognitive agent. Thus, we believe that, although this relationship is interesting in general, it is rather marginal when addressing the training of cognitive robotic agents with cooperating capabilities.
Comment 5: Conclusions contains no statement from authors, so the outcome of the entire analysis remains unclear.
Response 5: In the final section of the paper (Conclusions and future directions) it is stated that “The main challenge, in our opinion, is that generative AI cannot cooperate with humans because it does not really understand humans and thus is unable to support mutual understanding. The alternative roadmap suggested in this perspective essay is based on generative Artificial Cognition, i.e. an Internal Prospection machinery which is not imposed a priori from impersonal mega databases but emerges as a personal and personalized knowledge extracted from the personal interaction of a specific cognitive agent with a specific physical and social environment.” From a Perspective point of view, it is clear enough, in our opinion.
Reviewer 2 Report (Previous Reviewer 2)
Comments and Suggestions for Authors-
The reviewer has no further comments
Minor editing of English language required.
Author Response
Comment: The reviewer has no further comments
Response: Thank you for your appreciation.
Reviewer 3 Report (Previous Reviewer 3)
Comments and Suggestions for AuthorsThe paper presentation has significantly improved and now the paper merits publication.
Specifically, the paper is now divided into concise sections with descriptive titles that navigate the reader through this theoretical, perspective contribution.
The included figures are informative and describe the concepts described.
Finally, the text in the manuscript has been significantly improved to be clearer and to the point.
Author Response
Comment: The paper presentation has significantly improved and now the paper merits publication.
Specifically, the paper is now divided into concise sections with descriptive titles that navigate the reader through this theoretical, perspective contribution.
The included figures are informative and describe the concepts described.
Finally, the text in the manuscript has been significantly improved to be clearer and to the point.
Response: Thank you for your appreciation.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsI think authors uploaded wrong version - no formatting, no conclusions, no research ,just part of review and one figure.
Comments on the Quality of English LanguageEnglish seems acceptable.
Author Response
Comment 1: I think authors uploaded wrong version - no formatting, no conclusions, no research, just part of review and one figure.
Response 1: The article type is “Perspective”. Thus, no original research is required and there is no rigidly defined structure. However, in order to clarify the conceptual framework, the manuscript was modified by introducing a separation in the main conceptual points and the figure was subdivided in three parts accordingly.
Reviewer 2 Report
Comments and Suggestions for AuthorsThis paper focuses on collaborative robots with cognitive abilities, and also discusses ethical issues that may arise in emergency situations. This paper has some significance, but it still has the following problems.
1. The theoretical introduction of the cobots robot cognitive system is not complete enough, and there is no emphasis on highlighting the innovation of the cobots robot cognitive system. It is suggested to supplement relevant content.
2. The paper lacks experimental verification and does not conduct relevant experiments on the cognitive function of IR4 robots compared to IR2 and IR3 robots. It is recommended to supplement comparative experiments.
3. This paper lacks a detailed introduction to the current research status and progress of cobots robot cognitive function, and does not introduce the current status of cognitive function of related robots. It is suggested to supplement relevant examples. For example, it is recommended to cite references in the last paragraph of this paper on page 2: “Adaptive human robot interaction torque estimation with high accuracy and strong tracking ability for a lower limb rehabilitation robot, IEEE/ASME Transactions on Mechatronics, 2024, DOI: 10.1109/TMECH.2024.3394491”. This document provides a detailed introduction to the cognitive function and human-machine interaction of lower limb rehabilitation robots, which can serve as a supplementary explanation of the current research status. It is suggested to cite reference in the Developmental robotics section on page 3 of this paper: “Finite time observer based variable impact control of cable driven continuous manipulators, IEEE Transactions on Human Machine Systems, 2022, 52 (1): 26-40.” This literature provides a detailed introduction to the current research status and existing shortcomings of human-machine interaction in continuum robots. In addition, other literature is also available for reference. (a) Zhang C, Chen J, Li J, et al. Large language models for human–robot interaction: A review[J]. Biomimetic Intelligence and Robotics, 2023, 3(4): 100131. https://doi.org/10.1016/j.birob.2023.100131 (b) Chen L, Huang J, Wang Y, et al. Adaptive patient-cooperative compliant control of lower limb rehabilitation robot[J]. Biomimetic Intelligence and Robotics, 2024, 4(2): 100155. https://doi.org/10.1016/j.birob.2024.100155 (c) Liao Z, Chen B, Bai D, et al. Human–robot interface based on sEMG envelope signal for the collaborative wearable robot[J]. Biomimetic Intelligence and Robotics, 2023, 3(1): 100079. https://doi.org/10.1016/j.birob.2022.100079.
4. There are too many keywords in the article. It is recommended to cut down some of them and keep the article layered.
5. The number of images in the article is little. It is suggested to increase the number of images. For example, adding images of IR2 and IR3 as well as IR4 robotic cognitive system to increase its comparison and highlight the advantages of IR4 robotic cognitive system.
Comments on the Quality of English LanguageModerate editing of English language required.
Author Response
Comment 1: The theoretical introduction of the cobots robot cognitive system is not complete enough, and there is no emphasis on highlighting the innovation of the cobots robot cognitive system. It is suggested to supplement relevant content.
Response 1: The introduction was modified by clarifying a crucial aspect of robot cognition for effective human robot collaboration in IR4 and beyond: physical human robot interaction is only part of it, whereas prospection capability for autonomous behavior is the crucial issue (see also the response to comment 3).
Comment 2: The paper lacks experimental verification and does not conduct relevant experiments on the cognitive function of IR4 robots compared to IR2 and IR3 robots. It is recommended to supplement comparative experiments.
Response 2: The article type is “Perspective”. Thus, no original research is required and there is no rigidly defined structure. However, in order to clarify the conceptual framework, the manuscript was modified by introducing a separation in the main conceptual points and the figure was subdivided in three parts accordingly.
Comment 3: This paper lacks a detailed introduction to the current research status and progress of cobots robot cognitive function, and does not introduce the current status of cognitive function of related robots. It is suggested to supplement relevant examples. For example, it is recommended to cite references in the last paragraph of this paper on page 2: “Adaptive human robot interaction torque estimation with high accuracy and strong tracking ability for a lower limb rehabilitation robot, IEEE/ASME Transactions on Mechatronics, 2024, DOI: 10.1109/TMECH.2024.3394491”. This document provides a detailed introduction to the cognitive function and human-machine interaction of lower limb rehabilitation robots, which can serve as a supplementary explanation of the current research status. It is suggested to cite reference in the Developmental robotics section on page 3 of this paper: “Finite time observer based variable impact control of cable driven continuous manipulators, IEEE Transactions on Human Machine Systems, 2022, 52 (1): 26-40.” This literature provides a detailed introduction to the current research status and existing shortcomings of human-machine interaction in continuum robots. In addition, other literature is also available for reference. (a) Zhang C, Chen J, Li J, et al. Large language models for human–robot interaction: A review[J]. Biomimetic Intelligence and Robotics, 2023, 3(4): 100131. https://doi.org/10.1016/j.birob.2023.100131 (b) Chen L, Huang J, Wang Y, et al. Adaptive patient-cooperative compliant control of lower limb rehabilitation robot[J]. Biomimetic Intelligence and Robotics, 2024, 4(2): 100155. https://doi.org/10.1016/j.birob.2024.100155 (c) Liao Z, Chen B, Bai D, et al. Human–robot interface based on sEMG envelope signal for the collaborative wearable robot[J]. Biomimetic Intelligence and Robotics, 2023, 3(1): 100079. https://doi.org/10.1016/j.birob.2022.100079.
Response 3: most of references suggested by the rewiewer are related to the general issue of Human Robot Interaction, in particular physical interaction by means of different types of interfaces: haptic interfaces, EMG interfaces etc. Such interfaces involve adaptive control paradigms that require some degree of cognition or intelligence in a large sense. However, we believe that the crucial cognitive requirement for IR4 robots is prospection capabilities for achieving autonomous behavior and skill acquisition. The only exception is the paper by Zhang et al that proposes LLMs (i.e. mainstream AI) for human-robot interaction. We quote this paper, with a related additional one, at the end of the introduction, commenting that human robot interaction may help the development of cognitive tools but, by itself, is insufficient to achieve prospection capabilities essential for autonomous and collaborative performance.
Comment 4: There are too many keywords in the article. It is recommended to cut down some of them and keep the article layered.
Response 4: We reduced the number of keyword and re-structured the article in relevant sections.
Comment 5: The number of images in the article is little. It is suggested to increase the number of images. For example, adding images of IR2 and IR3 as well as IR4 robotic cognitive system to increase its comparison and highlight the advantages of IR4 robotic cognitive system.
Response 5: The original composite image was divided into three images, allocated in different parts of the article in order to clarify the key conceptual points of the proposed roadmap. For IR2 and IR3 robots well established software packages were developed, like the ROS operating system enriched by specific libraries, but they lack any significant true cognitive capability. We may also add that a lot of research activities were focused on the development of cognitive architectures for autonomous or semi-autonomous robots, via symbolic or sub-symbolic representation (Kotseruba and Tsotsos 2020, quoted in the submitted manuscript), but none of them succeeded to overcome the research level and influence IR4 and beyond in a significant manner. Thus no well established robotic cognitive system of IR4 is currently available and our proposal is not a specific system but a general bio-inspired roadmap to design a family of systems based on three main principles: 1) embodied cognition, 2) developmental robotics, 3) social robotics.
Reviewer 3 Report
Comments and Suggestions for AuthorsThe paper touches on a very interesting topic and provides an interesting literature review.
However, my recommendation is that the paper cannot be accepted for the following reasons.
The paper is not organised in sections, which makes it very confusing as to what is the introduction, related work, proposed approach, experimental results, and conclusions.
Besides the lack of organisation, there is no proposed contribution, nor an experiential evaluation of any method.
Figures should be placed appropriately in the text.
Perhaps, authors should reconsider submitting to a less technical and more philosophical journal that might be more appropriate given the nature of the proposed contribution.
Author Response
Comment 1: The paper is not organised in sections, which makes it very confusing as to what is the introduction, related work, proposed approach, experimental results, and conclusions.
Response 1: The article type is “Perspective”. Thus, no original research is required and there is no rigidly defined structure. However, in order to clarify the conceptual framework, the manuscript was modified by introducing a separation in the main conceptual points and the figure was subdivided in three parts accordingly.
Comment 2: Besides the lack of organisation, there is no proposed contribution, nor an experiential evaluation of any method.
Response 2: Again, the article type does not imply any experimental evaluation or experimental method.
Comment 3: Figures should be placed appropriately in the text.
Response 3: Thank you for the suggestion that was promptly adopted.
Comment 4: Perhaps, authors should reconsider submitting to a less technical and more philosophical journal that might be more appropriate given the nature of the proposed contribution.
Response 4: The roadmap for the design of cognitive agents capable to cooperate with humans is not a purely technical issue because it involves general problems, including ethical issues, that inevitably touch upon and border with a philosophical framework.
