Voronoi Tessellation for Efficient Sampling in Gaussian Process-Based Robotic Motion Planning
Round 1
Reviewer 1 Report
The paper addresses a very interesting topic in robotics, the generation of non-collision trajectories in dynamic conditions and unknown environments. The detection of obstacles is not a technical problem at this moment, but the effective generation of the trajectory raises serious problems, especially in dynamic conditions. The use of machine-learning type algorithms is fully justified for solving problems. The example for which the numerical results are presented seems simple, but this way of presentation does not affect the validity of the concept, in general. I recommend the authors approach more complex examples within concrete applications, such as those in the field of collaborative robots.
Author Response
Thank you for taking your time to review this manuscript. Please find the detailed responses in the attached file and the respective revisions in the re-submitted manuscript.
Author Response File: Author Response.pdf
Reviewer 2 Report
1- The paper needs language correction.
2- The abstract should be improved by removing the reference citation.
3- The literature survey is insufficient.
4- Divide the introduction into three subsections: 1) Literature review, 2) Research gap and motivation, and 3) Contribution and paper organization
5- Define all variables used in the equations
6- The X-axis and Y-axis labels and units are missing in almost all Figures
7- In conclusion,” While the proposed method is conceptually simple, however, the learning hyperparameters can be somewhat difficult to tune.”, which leads to constraints on the proposed method applications, State the limitation of the proposed method.
8- Rearrange the abbreviations alphabetic
9- Update references `
The paper needs language correction
Use Passive sentences (remove “we”)
Author Response
Thank you for taking your time to review this manuscript. Please find the detailed responses in the attached file and the respective revisions in the re-submitted manuscript.
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors address challenges in autonomous systems and claim to propose a novel approach called Voronoi tessellation sampling to improve Gaussian process-based imitation learning. This method aims to enhance data efficiency and scalability by selecting informative yet learnable data points based on spatial correlations. The authors reference a baseline experiment and claim that their approach achieves better policy learning with fewer data points across the entire feature space. Despite the contributions of the work are significant and standard, there are a few critical concerns that need to be addressed before publication:
1. The Introduction is written tangentially. There are many sentences where proper reference is required to improve the knowledge base of the manuscript. A few examples are 'Many conventional robotic systems have been used in static environment ... ... are feasible and adequate [?].' Moreover, the authors are suggested to explain more about existing works on static and dynamic motion planning of robots.
2. The authors are suggested to highlight the novelty of the work more carefully, as several works in the literature used the Voronoi tessellation method for dynamic motion planning. The contributions should be presented pointwise at the end of the Introduction section.
3. The notation 'w' in Eq. (1) should be checked with one mentioned before the equation. The notation Σ_y should be defined. Line 78, the word 'translateds' needs to be corrected. Please define \alpha from Eq. 6. In line 148, \beta should be symbolic in text.
4. At present, the results are minimally discussed. The authors are suggested to highlight the crucial values in Table 1 and then explain the relevant findings in more depth. The authors should explain the heatmaps in Fig. 8 to improve the paper's readability.
5. How the authors have tuned the hyperparameters? What is the value of \alpha from Eq. 6. The authors have used many small paragraphs that need to be merged for better readability.
Author Response
Thank you for taking your time to review this manuscript. Please find the detailed responses in the attached file and the respective revisions in the re-submitted manuscript.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
The authors have addressed all the concerns raised by the reviewer. However, for future, the authors are advised to mark the revisions in the manuscript for better understanding of newly added texts.