A Robust Disturbance-Rejection Controller Using Model Predictive Control for Quadrotor UAV in Tracking Aggressive Trajectory

Round 1

Reviewer 1 Report

Comments as attached.

Comments for author File: Comments.pdf

Minor editing/polishing is required.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper proposes a tracking control approach for a quadcopter that can track an aggressive trajectory using MPC and a nonlinear controller. The authors introduced motivation, described several procedures to design control law including visual expressions, and validated the performance of the proposed approach using several simulation results. However, overall, the quality of this paper needs to be improved, and this paper is not ready to be published based on the following comments.

- Overall, it might be better to clearly define the matrices, vectors, and scalar values with their dimensions in a proper manner. It’s difficult to follow the procedures with current definitions of the variables. Usually, a matrix, vector, and scalar variables are defined by a capital letter, a bold and non-italic small letter, and an italic small letter, respectively.

- In Eq. (5), do z_w and z_b indicate the unit vectors of the z-axis for the world and body frames?

- In Eq. (8), x and y are defined as the state and output variable, but the same letters in Eq. (34) are used as x and y positions. I recommend that the authors thoroughly define variables and derive all equations.

- In addition to this, x_c in Eq. (35) seems typo.

- In Section 4, the authors provided the information from the cited paper for the simulation parameters used in the simulation. However, it would be better to include those parameters in the manuscript for the reader’s convenience.

- In Section 4, the authors only provided the figures for the 3d trajectory, position, attitude, and angular velocity. It would be better to include the velocity and acceleration of the quadcopter as well.

- In Figures 3 and 4 in Section 4, the actual angular velocities are very inaccurate with several peak values, unlike the actual attitude, at the beginning of the simulation, which are undesired behavior. What are the reasons for having those peak values in the angular velocities during maneuver? The authors may need to include those analysis results in the manuscript.

- Figures 5(c) and 5(d) are the same. The authors need to include the proper figure.

- For the second case, the authors only include the 3D trajectory of the quadcopter. It would be better to include the state information like the other cases and include the analysis results.

- Some subsections mentioned the initial states and other parameters used in the simulation, but some other sections did not. It would be better to include that information for each case in a consistent manner.

- In Section 4.2.4, how do the authors include the state observation errors? Are the static errors added to the observation? If that’s the case, the authors may need to apply random values with the provided error magnitude for the realistic conditions. In addition, for the external disturbance, it might be better to include wind disturbance using proper wind models, instead of a sudden state change at a certain moment, to consider the realistic situation.

- Usually, MPC requires a lot of computational burden to find the proper solution as the authors mentioned, thus it may not be the proper approach for the real-time application. However, the authors claimed that the proposed algorithm can be solved quickly in real-time. However, I don’t find the results about the computational time, which prove that the proposed algorithm is applicable in real-time. Therefore, I recommend the authors include the computational time and its analysis result.

 

Typos

Line 17: QUAVs -> UAVs

Line 196: . (period) -> , (comma) after the reference [27]

Overall, there are several grammar errors in the manuscript. The authors need to thoroughly review the manuscript.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Referee report concerning “A Robust Disturbance Rejection Controller Using Model Predictive Control for Quadrotor UAV in Tracking Aggressive Trajectory” by Z. Xu, L. Fan, W. Qiu, G. Wen and Y. He

 

The paper proposes a control scheme to realize a waypoint tracking of a quadrotor. In the opinion of this reviewer, the paper contains a series of unclear expositions and misuse of terms. The authors should pay serious attention to the following points:

1.- First, why do you call a waypoint tracking if you use a continuous curve?

2.- I do not see how the discrete-time model (5) corresponds with the quad model, many constants are missed.

3.- Important details of the MPE approach to control are missed too like the time horizon in which the problem is working, and how this connects with the rest of the controllers. Does the nonlinear control have a time limit?

 

4.- Using the sign function does not mean a sliding mode control is being developed, where is the reduced dynamic manifold defined? How does the term hmax ensure compensation? This is not clear!

5.- A real-time experiment would increase the workability of the approach.

The language quality should be improved.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Most of the problems are addressed in the author's response.

N/A

Reviewer 2 Report

The revised manuscript properly reflects the comments and is ready to be published. 

 

There are some minor grammar errors. The authors need to review the manuscript thoroughly and polish it.