Self-Assembly and Self-Repair during Motion with Modular Robots
Round 1
Reviewer 1 Report
The article provides novel self-assembly method, repair strategies, and data structure for modular robotic structures.
The presented methods are well presented and understandable.
Some of the figures contain small text which makes them hard to read.
Author Response
Please see attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The authors have presented self-assembly and self-repair techniques for a modular self-reconfigurable robot system (MRS). The proposed techniques are validated in simulation and on real robots.
The attached videos are nice but in my opinion, they are too long. Please create more concise videos (maybe 1 each for self-assembly and repair), which will capture the main ideas of the paper. Potentially, speed up the videos, otherwise, they seem too slow.
Specific Comments:
- The self-assembly technique (at least in the static state) is not compared with any previous approach in MRS. Can we see some comparison results even with a greedy kind of algorithm?
- There are a few relevant references that I know of in self-assembly that is missing from the paper (see below). How does your proposed approach differ from theirs? How do you advance the state-of-the-art from theirs?
- The paper is currently organized poorly, in my opinion. There is a lot of text but less algorithmic structure. Please provide pseudo-codes of the proposed algorithms for better readability and then refer to them in the text. Otherwise, the paper is too long at this point.
- For self-repair, how long does a module wait before understanding that the neighbor is not operational anymore? If we vary this time, is there any change in the performance of the self-repair technique?
References:
- Li, H., Wang, T., & Chirikjian, G. S. (2016). Self-assembly planning of a shape by regular modular robots. In Advances in Reconfigurable Mechanisms and Robots II (pp. 867-877). Springer, Cham.
- Dutta, A., Dasgupta, P., & Nelson, C. (2019). Distributed configuration formation with modular robots using (sub) graph isomorphism-based approach. Autonomous Robots, 43(4), 837-857.
Author Response
Please see attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
This paper deals with self-reconfigurable modular robots that are able to combine into more complex structures in order to perform certain tasks.
The introduction of the article contains a detailed overview of the current state in this area and specific references to references of similar works of other research institutes. An overview of previous work in this area of research is given in great detail.
The authors solve reconfigurability strategies and even solve the problem of faults that can occur in the modular arrangement of robots. This is an interesting concept of a swarm of robots that can be operationally combined into various forms of organization. Conceptually, this proposal is interesting, even from a technical point of view. The authors use simulation tools to verify the proposed strategies, the results of which are documented in this article. The results are remarkable and represent a unique solution for reconfigurable robots. The authors also provided supplementary materials in the form of video files, which document the outputs of the simulations, and experiments with functional prototypes of robots are also interesting. These videos show that this is a viable proposal that can be further developed.
The main contribution of this work is the creation of a system of self-assembly of a modular robotic organism, the creation of a new data structure for the description of the modular structure and the design of a strategy enabling both self-repair and self-assembly.
The article is very well technically processed and the individual results are correctly documented.
At the end of the article there is a very detailed discussion of the achieved results and also future research work is presented.
Comments:
The authors refer to their previous work, where they dealt with the structural design of the robot, but even so, it would be good to state the kinematic structure of one robotic module, where the structure of the robot and the docking system of connecting robots during the motion would be clear.
References to conference articles do not have complete bibliographic data (conference location, conference date, pages, publisher, volume, etc.), check the correctness of other references as well.
Author Response
Please see attachment
Author Response File: Author Response.pdf
