A Simple Soft Computing Structure for Modeling and Control

Round 1

Reviewer 1 Report

1. The overall structure of the control system is not clear. From reading this paper, it is difficult to know where and how the developed computation model can be placed for a control system. For instance, does the approximate model in Fig. 3 denote the developed model given in Fig. 4? If yes, a detailed block diagram should be given. 2. In Fig. 3, what do the system, kinematic block, and approximate model represent? 3. It cannot see the relation between equations (5a)-(5f) and Fig. 1, and what did the authors take equations (5a)-(5f) for? 4. What is the difference by using Figs. 1 and 4 for modeling and control? 5. All information about the networks in Figs. 1 and 4 should be given, including the functions of neurons, layers, connection structures, and weights.

Author Response

"Please see the attachment."

Author Response File: Author Response.pdf

Reviewer 2 Report

Very interesting idea and very nice explained. I have only few formal observations:

1) When we refer to a trajectory described by a point in Carthesian space we think to a curve, and in case of circular motion a closed curve. So instead of "amplitude" (pay attention that A is used for a vector) and "circular frequency" (better to use small letter "omega" from Greek alphabet) is maybe much proper to use radius of circular motion (R is used too ...) and angular speed (with unit [rad/s]) respectively. This observation comes from my major which is Robotics not control !

2) In row 199 (which is in fact a paragraph ...), right above the relationship (2) is wrote "... [q] m and [Q] N ..." and the right brackets seems to be used reversed.

3) Increase the size of Fig. 5 cases because is difficult to read text and numbers. Exchanging the positions of lower cases (c) and (d) between them is maybe better in order to compare with the cases (a) and (b) located up to them.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

1) The introduction is too long and should be refined.

2) Why chose “the van der Pol Oscillator” as the teaching Example? Do other nonlinear systems have the same effect?

3) The authors claimed that the new neural model provided satisfactory results. What’ s the satisfactory results? Please give some detail explanations. Small error? Fast convergence? Low computational complexity?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

All comments have been addressed.

Reviewer 3 Report

The authors revised the manuscript according to the reviewer's comments, so I recommend it for publish.