Compliant Detachment of Wall-Climbing Robot Unaffected by Adhesion State
Round 1
Reviewer 1 Report
The paper aims at proposing a strategy to control climbing robots. The idea is interesting and the experimental results sounding. I suggest to accept the paper after a minor revision solving the following issue:
The role of the vision system appears to be critical since it provides the actual adhesion state. Authors should clarify this point by adding details on the image processing techniques used: are reliable in realtime? (see e.g. Gagliano, S., Stella, G., & Bucolo, M. (2020). Real-time detection of slug velocity in microchannels. Micromachines, 11(3), 241.).
Author Response
Please also find the attachment.
--
Response to Reviewer 1 Comments
Point 1: The paper aims at proposing a strategy to control climbing robots. The idea is interesting and the experimental results sounding. I suggest to accept the paper after a minor revision solving the following issue:
The role of the vision system appears to be critical since it provides the actual adhesion state. Authors should clarify this point by adding details on the image processing techniques used: are reliable in realtime? (see e.g. Gagliano, S., Stella, G., & Bucolo, M. (2020). Real-time detection of slug velocity in microchannels. Micromachines, 11(3), 241.).
Response 1: The image processing system is used for data acquisition of adhesion area and does not serve as the feedback of the control system, so it ran offline. The calculation of the adhesion area was mainly realized by the algorithm of frame difference and binarization (see Figure A1). The algorithm could run at a frequency of 30 Hz on our laptop (i7-9750) and can also be used for real-time processing. The suggested reference has been added as [28]. More details have been updated to the revised version:
Adhesion area test platform and adhesion area detection process. The adhesion area actual image captured by the camera is uploaded to PC for image processing [28]. The frames containing only markers (non-adhesion) were recorded as fbg, and the frames after adhesion were recorded as fsrc. Through the difference between the two frames (Frame difference), the frame fad containing only adhesion region was obtained, which was binarized and the total number of pixels of adhesion region in the frame was calculated (Binarization). The actual adhesion area is obtained by the ratio of the number of pixels of the reference object to the actual length.
[28] Gagliano, S., Stella, G., & Bucolo, M. (2020). Real-time detection of slug velocity in microchannels. Micromachines, 11(3), 241.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper presents a control system for a wall-climbing robot. The paper is interesting and easy to read. The manuscript has been improved with respect to the previous submission.
However, no quantitative evaluation of the vibrations on the prototype is shown. It would be interesting to discuss how the vibrations on the robot are (or can be) measured during the motion of the experimental prototype.
A block diagram of the control system should be added to the manuscript. It would be interesting to discuss how the controller has been tuned.
The quality of some of the graphical plots is very low. I suggest using figures with higher quality and resolution.
Author Response
Please also find the attachment.
--
Response to Reviewer 2 Comments
Point 1: The paper presents a control system for a wall-climbing robot. The paper is interesting and easy to read. The manuscript has been improved with respect to the previous submission.
However, no quantitative evaluation of the vibrations on the prototype is shown. It would be interesting to discuss how the vibrations on the robot are (or can be) measured during the motion of the experimental prototype.
Response 1: In the field of quadruped robots, the vibration of robot always be measured by the amplitude of the attitude angle (RPY). In this paper, because of the great variance of adhesion, we evaluate the vibrations by standard deviation of the rolling angle which can best reflect the change of normal force during the experiments (see Figure 5 (b) and Figure 7). In Section 4 Robot Stability on Various Surfaces with Various Slope Angles, the stabilities of TTC and OIAC are measured in this method. More information has been updated to the revised version:
To evaluate the motion stability of the robot, the roll-pitch-yaw (α-β-γ) attitude information and the climbing failure rate of IBSS-8 on the glass and Teflon surfaces at various slope angles were recorded several times. The standard deviation of rolling angle α which can best reflect the change of normal force was calculated and used as the quantitative evaluation of the vibrations. The statistical results are shown in Figure 7.
Point 2: A block diagram of the control system should be added to the manuscript. It would be interesting to discuss how the controller has been tuned.
Response 2: A new motion control strategy diagram and more details have been updated in Figure 2(b):
The expected foot end coordinate of the single limb is given by the trajectory. The expected angle and angular velocity of the joints are obtained through the inverse kinematic of the single limb, and the feedback of the current angle and angular velocity is obtained. The error e is input to online impedance adaptive controller, and the joints torque are output to control the single limb.
Point 3: The quality of some of the graphical plots is very low. I suggest using figures with higher quality and resolution.
Response 3: All dpi of graphical plots has been updated to 300 in revised version.
Author Response File: Author Response.pdf
Reviewer 3 Report
The paper presents a method to improve the stable movement ability of the wall-climbing robot. It is a theoretical and also experimental work, finalized with logical and justified conclusions .
Line 64- I suggest to add a visual scheme to Kendall model expressed by eqs.1 and 2.
Author Response
Please also find the attachment.
--
Point 1: The paper presents a method to improve the stable movement ability of the wall-climbing robot. It is a theoretical and also experimental work, finalized with logical and justified conclusions.
Line 64- I suggest to add a visual scheme to Kendall model expressed by eqs.1 and 2.
Response 1: Fig.1 has been added to the revised version to describe the Kendall model.
(Figure, please find attachment).
Author Response File: Author Response.pdf
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
- The basic information such as the size of each link of the robot, the joints of the links movable area, and the position of the center of mass of the robot are not shown, I cannot evaluate the paper. It is necessary to describe the model and to describe all parameters.
- Some mathematical expressions are shown, but the paper does not show the model of the robot, so it is difficult for the reader to evaluate them.
- Figure 3 is supposed to explain the movement for one leg, but I think the leg of this robot cannot move the link to the direction.
- The figures in the paper have many tiny and obscure letters and numbers. The author needs to revise all of them significantly. The one graph in Figure 5 do not have description of the axes.
- What do the orange lines mean in the paper? It's very hard to read. You could deleted the lines.
Reviewer 2 Report
The paper presents a strategy for the compliant detachment of a wall-climbing robot with controllable adhesion strength. The paper is interesting and easy to read. The method and the results are properly described and commented. However, the following points need to be considered before publication.
- It would be interesting to briefly compare the adhesion method with other approaches for wall-climbing robots (magnetic attraction, vacuum suction cups, gripping claws). Some suggested references are reported below.
- No quantitative evaluation of the vibrations on the prototype is shown. It would be interesting to discuss how the vibrations on the robot are (or can be) measured during the motion of the experimental prototype.
- It would be interesting to discuss the future possible autonomous motion and path planning (without an external controller) of the robot to accomplish a task independently from the human input.
Minor comments:
- The autonomy of the batteries of the robot should be reported in the paper.
- Please check the formatting of Equation (4) which is not in line with the text.
Ge, D., Tang, Y., Ma, S., Matsuno, T., & Ren, C. (2020). A pressing attachment approach for a wall-climbing robot utilizing passive suction cups. Robotics, 9(2), 26.
Eto, H., & Asada, H. H. (2020, May). Development of a Wheeled Wall-Climbing Robot with a Shape-Adaptive Magnetic Adhesion Mechanism. In 2020 IEEE International Conference on Robotics and Automation (ICRA) (pp. 9329-9335). IEEE.
Seriani, S.; Scalera, L.; Caruso, M.; Gasparetto, A.; Gallina, P. Upside-Down Robots: Modeling and Experimental Validation of Magnetic-Adhesion Mobile Systems. Robotics, 2019, 8, 41.
Liu, J.; Xu, L.; Chen, S.; Xu, H.; Cheng, G.; Xu, J. Development of a Bio-inspired Wall-Climbing Robot Composed of Spine Wheels, Adhesive Belts and Eddy Suction Cup. Robotica 2020, 1–20.
