Human-Robot Interaction Torque Estimation Methods for a Lower Limb Rehabilitation Robotic System with Uncertainties

Round 1

Reviewer 1 Report

In this manuscript, Yepes et al. developed two human-robot interaction torque estimation methods which both show better accuracy than the approaches reported in the literature. Overall, the motivation and novelty of this paper are very clear, the problem and model developments are nicely presented. The reviewer can recommend for publication with the following minor comments.

It is understandable that this paper is mainly simulation based, however, it would be good if the authors could discuss the future work in applying to their robotic system (e.g. what else is needed, what the challenges will be).

In Figure 7 (a), what is the cause of the high noise level in the beginning (t < 1 s)?

Author Response

We would like to thank the insightful comments provided to improve the paper.

Next, we will present a point-by-point response to the referees’ comments. Moreover, we attach a marked-up revision to the manuscript using the “Track Changes” function.

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript presents a rather theoretical contribution to the complex issue of a lower limb rehabilitation, assisted by robotic systems. This is a very important topic, especially considering the rising number of armed conflicts all over the world. The focus on cost reduction by excluding certain expensive sensors is a welcome aspect of the manuscript, too. 

While the manuscript is quite comprehensive, there is some information I would like the authors to add: 

1. In section 2 (Requirements definition) you mention a survey and a literature review conducted in order to define requirements of your human-robot interaction torque estimation method. Please, provide links to papers or documents describing the survey and the literature review in more detail. If no such links can be provided, please specify at least the questions used in the survey and answers to them in an exact way. 

1. Please, describe your plans or vision for the practical implementation of the achieved results in the conclusion of the manuscript. 

In addition, English should be checked as there are some typos and cases of incorrect spelling or repeated words. There are also several places where the language is a bit naïve in style. 

Author Response

We would like to thank the insightful comments provided to improve the paper.

Next, we will present a point-by-point response to the referees’ comments. Moreover, we attach a marked-up revision to the manuscript using the “Track Changes” function.

Author Response File: Author Response.pdf

Reviewer 3 Report

1. The abstract needs to be written in a more clarified way indicating the contribution and the superior of one proposed method to other.
2. The introduction section has redundant text which has more generic information. This applies to section 3.
3. Other recent and relevant works which belong to Lower-limb or observer methodology has to be added. I suggest (Robust Adaptive Control of Knee Exoskeleton-Assistant System Based on Nonlinear Disturbance Observer), (On the design of backstepping controller for Acrobot system based on adaptive observer).
4. In the introduction section, there is an extensive explanation of the Nukawa robot, while the authors stated that the effectiveness of estimators has been evaluated based on numerical simulation.
5. The title modeling section is converted to a Dynamic model since the modeling has not been developed but referred to.
6. Some abbreviations have to be extended where they are mentioned for the first time.
7. The authors have to explain why the observer (based on Saadatzi) has been chosen.
8. The establishment of Eq. (6) is not clear.
9. The transformation process of the model is not clearly presented.
10. The authors mentioned that "Shimkin et al. [44] to accomplish the persistent excitation". This phrase has to be extended.
11. The article is mainly developed according to other research, which is spread alongside the manuscript. Where is the novelty?
12. The stability analysis of suggested observers has not been addressed.
13. What is the principle of separation for an observer-controlled system of this study?
14. Eq. (19) has been mentioned without proof or citation.
15. The results have to focus on the actuating torques. Is this study has addressed the saturated input torque or considered the unbounded torque?
16. The authors have not addressed the boundness of estimated parameters and the upper bound of uncertainties.
17. The practical realization of proposed observers has to be implemented within a real environment. The authors have started as if the robot is actually applied!!
18. The conclusion is lengthy and descriptive and it has been abbreviated and the concluded points have to be clearly stated. In addition, the improvement percentages have to be numerically reported.

Author Response

We would like to thank the insightful comments provided to improve the paper.

Next, we will present a point-by-point response to the referees’ comments. Moreover, we attach a marked-up revision to the manuscript using the “Track Changes” function.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

I appreciate the great efforts that the authors have made in response to my questions and comments.

All my comments have been addressed—no further comments.