Design, Control, and Testing of a Multifunctional Soft Robotic Gripper
, Afonso Leite 1,2, Francisco Campos 1,3, Nuno Monge 4, André Rocha 4 and Mário J. G. C. Mendes 1,2,5,*Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis research developed and tested a multifunctional soft robotic gripper using pneumatic network (PneuNet) bending actuators, suitable for handling sensitive products. The study explored various fabrication methods including 3D printing and molding, employing materials like FilaFlex, Elastosil M4601, and Dragonskin Fast 10, and verified their functionality through simulations and experimental validations. Here are my comments:
The control strategies (PI and On-Off controllers) are described superficially without detailed implementation specifics or discussion on why these were chosen over other potential strategies. How do these controllers compensate for the nonlinearities and hysteresis characteristics of soft actuators?
Soft actuators often have slower response times and reduced precision compared to hard actuators. How do these limitations affect the performance of the robotic gripper in practical scenarios?
Standard deviations, error bars, or confidence intervals in graphical representations are missing. Recommend to include a more comprehensive analysis using statistical tools like ANOVA or regression analysis to discern the influence of different design and control parameters on the actuator's performance.
This paper lacks a recent related advancements, like Controlling a peristaltic robot inspired by inchworms; and Novel soft robotic finger model driven by EHD pump.
Please explain the Finite Element Method (FEM) simulations used in your study. What assumptions were made in your models?
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors propose a multifunctional soft robotic gripper for manipulation.
The topic is of interest for academic and industrial research, as the existing literature shows.
When comparing the different solutions, the authors should provide additional details related to the proposed application. In particular, even if there are no specific analyses yet, as discussed in the conclusion, some estimates of the power consumption, payload and lifetime should be provided for each actuator compared to the others.
In addition to the standard PID-based controllers, the authors should also mention recent work presenting “a new embodied motor-neuron architecture” with potential applications for robot control by using “a nullcline-based control strategy for PWL-shaped oscillators”. These alternative solutions, which aim to reduce the complexity of the control network, should be discussed.
The results shown in Fig. 13 do not seem to match the angle shown in Fig. 11, where 115 degrees are achieved at 1 bar, while more than 120 degrees are achieved at 0.4 bar in Fig. 13.
The quality of some figures (e.g. Fig. 21) should be improved.
Typos and problems with spacing in the manuscript should be corrected.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsReview
The paper here presents the design, control, and testing of a multifunctional soft robotic gripper designed for a Dobot robot, using pneumatic network (PneuNet) bending actuators. The authors explore two PneuNet models fabricated via 3D printing and molding and test three materials—FilaFlex, Elastosil M4601, and Dragonskin Fast 10. The paper includes both simulation results and experimental testing to validate the functionality of the gripper, focusing on actuator bending behavior, manufacturing techniques, and control methods.
Overall Evaluation
My overall impression on the paper is that the authors make a good effort to build upon existing designs, develop control schemes, and model the behavior of the actuators. However, there is large room for improvement on organization of information, critical evaluations of theory and designs and attention to details. The work attempts to recreate or build upon the designs of actuators from two different papers, with 3 different materials, and 2 different methods. In addition, although the work is complete in its goal with experiments and simulations, I find the work slightly lacking in novelty as it focuses on recreating/improving existing designs (perhaps the design is indeed quite different? The authors use the words “inspire” but they have not given details about how novel these designs are from their references. From a basic reading, it does not seem very different) with two 3D printing and molding (which are quite common techniques), with different yet common molding materials. As mentioned, the theoretical explanations, writing, organization can be significantly improved. Please refer to the following sections for details.
3. Major Comments (Essential Revisions)
In section 2.1, the author mentions two different models. Which one is used for the simulations? From my understanding, the simulation does not use either of the models. Please justify the use of these two models here and how they are applied, if any of them are applied. If the information is given in the work, it is not very clear at all. The simulation simply refers to the hyper-elasticity models. Also, if they are needed, please put up diagrams for the second model, which has no graphic representation at all.
For 2.2, it seems redundant to separate into multiple figures just for showing the products in different materials and molds; and type II is shown in one of the figures and not the others. The information can be presented in a more concise, thoughtful manner. For tables in this section, a single one would suffice with different materials at different columns.
I have no comments on the controller, as I have no prior experiences on this subject.
In 3.1, author states “As Yap et al. [18] noted, the 3D printed elastomer can have an anisotropic behavior. 286 In the present work it was assumed that its behavior was isotropic.” Is there further justification for this assumption, as literature explicitly states the opposite? What impact this may have? This needs further comments.
In 3.2, two models are used. Is there a reason why different models are used for the molded polymers and the printed ones? Please make further comments. Also, in line 297, it says 3rd order mooney, but in the table, it says 2nd order. Which one is correct?
In 3.3, accurate predictions were made. However, the matching is only correct at the ends. The intermediate pressure ranges show quite an off trend. Experiments are sublinear, while simulations are super linear. Also, why is linear regression used on this data? The models and experiments are clearly nonlinear. Many of these are unanswered or not explained at all. Same with type 2, and the off trend in type 2 is also unexplained.
Again, the plots of simulations and the pressure vs angle plots should be organized into one or two plots.
Recommendation
The paper tries in designing and testing grippers with different methods of manufacturing, and the design of the controller. However, there is no new distinction on what is new in their designs. It is not clear why making them via 3D printing and molding (which are the primary methods for most soft robotics manufacturing) is novel. The materials used are common and fairly well tested it seems, as data on their material properties and hyper-elasticity are well documented for simulations. If the work is indeed novel in all these aspects, the author has not explained it well or attempted to present critically to the audience. Organizations need to be significantly improved; and significant explanations need to be given for recommendation of publications.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe revised paper looks much better now.
Reviewer 2 Report
Comments and Suggestions for AuthorsThe paper has been properly revised
Reviewer 3 Report
Comments and Suggestions for AuthorsThe author has thoughtfully included revisions that answer most of the pressing concerns addressed in previous review.
