Temperature Estimation of Thin Shape Memory Alloy Springs in a Small-Scale Hip Exoskeleton with System Identification and Adaptive Control

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript titled “Temperature Estimation of Thin Shape Memory Alloy Springs in a Small-Scale Hip Exoskeleton with System Identification and Adaptive Control” presents the design of a small-scale hip exoskeleton, establishes system transfer functions, and proposes a temperature estimation method. The study addresses a cutting-edge topic, but the current version has several shortcomings and requires revision.

1. 1. In the abstract and introduction, the authors should refine the sentences to help readers better understand the research objectives. While the abstract summarizes the work, the innovative aspects are not sufficiently highlighted. For example, the “temperature estimation algorithm” is the core of the study, but its differences and advantages compared with existing SMA temperature measurement methods (such as resistance-based or infrared methods) are not clearly presented.
   2. On page 5, line 148, the label “Figure 3” should be consistent with other references (e.g., Fig. 4). In addition, references to Fig. 7, Fig. 8, and Fig. 9 appear to be missing in the main text; it is recommended to check and include them.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Improvements to the language and syntax.
Optimized figure citations. For example, "Figures 8" and "Figures 9" on page 9 are not referenced in the text.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Summary

The manuscript proposes a small-scale hip exoskeleton actuated by thin Shape Memory Alloy (SMA) springs arranged in a bi-directional/antagonistic configuration. The work integrates system identification, PID/adaptive PID control, and a temperature estimation method to address the challenge of measuring SMA spring temperature. The study provides experimental validation including thermo-mechanical characterization, system identification, and control experiments (stepping and sinusoidal trajectories). The results demonstrate reliable trajectory tracking (RMSE ≈ 0.94°) and effective temperature estimation. The paper addresses an important and relevant topic in wearable robotics and demonstrates technical depth.

General Comments

1. Clarity of Temperature Estimation Method

The paper introduces a computational temperature estimation model based on heat transfer and voltage/current data. While the description is mathematically correct (Eqs. 12–15, p. 10), the implementation details are not always clear. For instance, how robust is the method to noise in current/voltage sensing, and how is calibration against thermocouple measurements handled?

Recommendation: Please provide a more detailed discussion of the robustness and limitations of the estimation method.

1. Control Strategy

The adaptive PID control is described and validated (Figs. 14–18, pp. 14–18). However, the choice of learning rates (γ1, γ2, γ3) and initial PID gains appears ad hoc, and no systematic tuning methodology is provided.

Recommendation: Clarify whether these values were obtained experimentally by trial and error, or through a structured method.

1. Experimental Validation

The characterization experiments (Fig. 13, p. 13) clearly demonstrate flexion and extension, but the need to reduce the stroke range due to overheating is a key limitation. While acknowledged, this issue deserves more discussion, particularly in terms of long-term durability and safety for wearable applications.

1. Novelty and Contribution

The manuscript’s main contributions are (i) the dual antagonistic SMA spring arrangement for hip actuation, (ii) the temperature estimation algorithm, and (iii) the adaptive PID control applied to the hip joint. These points are clear but could be emphasized more explicitly in the conclusion.

Specific Comments

• Abstract (p. 1): The contribution of the temperature estimation algorithm should be highlighted more clearly.
• Section 4, Control Algorithm (pp. 7–8): The block diagrams (Figs. 6–7) are helpful but would benefit from numerical details (sampling rate, control frequency).
• Section 5.1, Temperature Estimation (pp. 8–10): The RMSE values in Figs. 10–12 are encouraging, but please discuss whether they are sufficient for safe operation (i.e., maintaining SMA below damaging thresholds).
• Figures 14–17 (pp. 14–17): The results are convincing, but axes could be labeled with clearer units and ranges for easier interpretation.

Conclusions

This manuscript presents relevant and novel contributions to the field of SMA-driven wearable robotics. The work is scientifically sound and well structured, with convincing experimental results. The main weaknesses relate to the clarity of control tuning and temperature estimation robustness. Addressing these points will significantly strengthen the paper.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Revision is required to improve the quality of this work.

• guiding the tuning of PID controllers

Full term should be provided before short term is used. Pls check throughout the whole paper.

• The common shapes for SMAs are wire, spring, ribbon, and Clydesdale.

Clydesdale? Pls use commonly used terms in technical write-up.

• Eqn 2 is an oversimplified model without considering the length of SMA wire and boundary conditions. Refer to

A note on size effect in actuating NiTi shape memory alloys by electrical current, Materials & Design, 29, 2008, 1432-1437

for details of size effect in Joule heating of Nitinol wires.

• where ΔL is the linear movement of part (4) caused by the SMA springs

part (4)?

• As given in the heat transfer model in Eq.(3)

Eq.(3)? Pls double check the whole paper to ensure the quality of the paper.

• Figure 14.

All symbols in the figures must be properly defined. Same for other figures.

• Specifics of the SMA springs must be provided, testing conditions should be mentioned.
• More experiments are required to verify the reliability of the system, e.g., different ambient temperatures.
• The response time is slow. How to improve it?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have made sufficient edits and answered all questions/comments clearly. We recommend the manuscript for publication