Computational Simulation of an Agricultural Robotic Rover for Weed Control and Fallen Fruit Collection—Algorithms for Image Detection and Recognition and Systems Control, Regulation, and Command

Round 1

Reviewer 1 Report

Major comments:

1. The authors should clarify if there is any scientific novelty in their study. So far, the poor introduction section, which contains only three references to the master’s theses and zero ones to any peer-reviewed publications, does not allow estimating the authors’ contribution from a scientific point of view.
2. Page 3, Figure 3. It would be beneficial to provide numeral positions in the figure and add a corresponding description in the text for a better understanding of the rover design. For example, line 103 mentions a Cartesian arm, but it is hardly seen in the figure.
3. Throughout the paper, the authors mention x, y, and z directions and axes several times (for example, p. 4, l. 132–133; p. 5, l. 142; p. 7, l. 222; p. 10, Fig. 12; p. 13, l. 349–352). The authors, however, do not describe these directions and axes explicitly in the text.
4. P. 5, Fig. 5. It would be beneficial to explicitly designate joint axes to clarify the gripper kinematics (maybe in a separate kinematic scheme near the figure). Later, in Fig. 12, the authors refer to joints no. 1 and 2, main gripper axis, and “0” position, but these notations are difficult to understand at the moment.
5. P. 7, Fig. 6. The figure has numerical positions in the red boxes, but there is no corresponding description in the text (I suppose these positions relate to the bulleted list on p. 6).
6. P. 7, l. 225. The authors write, “It was chosen to use only two of the joints of each gripper,” but provide no explanations. Moreover, it is unclear what the authors mean by “each gripper” (How many grippers are there? Does the “gripper” mean one finger or the entire end-effector?) and what joints they choose (see also comment #4).
7. P. 8, Fig. 8. The figure has poorly seen notations in the red boxes undescribed in the text (see also comment #9).
8. Subsection 2.3. About the proposed algorithm:
   • The authors consider the rover motion only between two tree rows. What will happen when this task is completed? Is there any system that performs the rover path planning around the entire orchard?
   • Is it possible to pick fruits and spray weeds that stay outside the rover width or right under its wheels?
   • Is there any control system that tracks the rover motion? For example, what will happen if the rover suddenly goes off its course and starts moving towards the trees?
   • How does the gripper know it is holding an object? Are there any force sensors in the gripper?
   • Fig. 11. The flowchart has a decision block named “Can gripper capture object?” What happens to the objects the gripper could not capture?
   • Fig. 12. It is difficult to imagine the picking procedure and gripper movements from the figure (see also comment #4). It would be beneficial to provide an additional figure (similar to Fig. 17) to clarify the picking steps.
9. The authors should improve the quality of all the figures: at the moment, they look blurry when zooming.

Other comments:

10. P. 4, l. 127. The authors write, “Currently about 40% of food loss is caused by pests, pathogens and weeds.” The authors should provide a reference that proves this statement.
11. P. 5, Fig. 4. The authors should explain why there are two grippers in the figure or remove any. Do these two grippers correspond to their extreme positions on the rail? If so, the text states nothing about this.
12. P. 5, l. 142. The meaning and location of “spacer” are unclear (as well as “y1 axis,” see also comment #3).
13. P. 8, Fig. 9. It is difficult to see the camera, mentioned in l. 243, in the figure (see also comment #9).
14. P. 11, l. 275. What does the asterisk after “measurements” signify?
15. P. 11, Fig. 13. The authors should translate “Linha” and “Enterlinha” into English or remove them from the figure.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This work investigates a computational simulation of an agricultural robotic rover for weed control and fallen fruit collection. The work is interesting and well-structured, but it needs to cope with some concerns as follows:

1.In the Introductory section, the structure of the paper should be given.

2.In the Introductory section, some development of wheeled robots in recent years should be given, especially in the improtant field of tracking control of wheeled robots (please refer to the literature: adaptive sliding mode based disturbance attenuation tracking control for wheeled mobile robots, extended state observer based adaptive sliding mode tracking control of wheeled mobile robot with input saturation and uncertainties), so as to enrich the background of the article.

3.The picture should be clearer, for example Figures 6, 11, 12.

4.The research significance or importance of this paper should be emphasized.

5. A detailed analysis of the contributions or highlights of the paper should be given.

6. In recent years, drones have also had prominent advantages in the field of agricultural sowing. In the future, whether the method designed in this paper can be combined with uav for collaborative operation. Therefore, it is suggested to give the application of UAV in the conclusion section (please refer to anti-saturation adaptive finite-time neural network based fault-tolerant tracking control for a quadrotor UAV with external disturbances, Fixed-time disturbance observer-based robust fault-tolerant tracking control for uncertain quadrotor UAV subject to input delay, Antisaturation finite-time attitude tracking control based observer for a quadrotor)

7. Some literature should be added to enrich the background of the article

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have addressed all the comments and improved the quality of the paper. There are only a few points to check:

1. The authors have extended the introduction section. Most of the added material, however, is irrelevant to the topic considered in the paper. For example, there are new paragraphs that describe robot dynamics and control and mention serial and parallel robots (references [8–29]). This material seems unnecessary here because it has almost nothing to do with a rover simulation considered by the authors. In contrast, references [30–35] are relevant and complement the research appropriately.
2. Pages 5–6, lines 200–201. Δx and Δy dimensions do not correspond to the axes directions and working envelope given in Fig. 4.
3. P. 9, Fig. 8. The right fragment of the figure shows a red axis with an empty red box. What is the purpose of these notations?
4. The quality of the figures is still low: the figures look blurry when zooming. The authors should increase the resolution of their figures (600 dpi at least) and disable image compression.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf