Path Planning for Mobile Robots Based on a Hybrid-Improved JPS and DWA Algorithm

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors in the paper “Path Planning for Mobile Robots Based on a Hybrid Improved JPS and DWA Algorithm” effectively explain how the traditional Jump Point Search (JPS) differs from the standard A* algorithm. They improve path planning performance for mobile robots in complex environments. In global planning, a quadrant pruning strategy guided by the target direction and a sine-enhanced heuristic function reduces the search space and accelerates planning.

For the benefit of the paper, I have few suggestions

1.  In Eq. (3), sin(α/β) is introduced without defining α and β clearly in mathematical terms (are they scalars or vectors, and are α and β angles or parameters?).

And why is chosen sin function heuristics?

2. While for JPS metrics (time, number of nodes) are shown, for DWA the section is more qualitative. Although the text explains what is being improved, but there is no table or graphic showing the effects of the new heuristics or route prescription, for example: Reduction in number of nodes, reduction in computation time, Improvement in path length
3. No computational cost analysis The algorithm is for real-time use, but it is not stated how fast it runs, think to add time analysis (fps or time per iteration)
4. It would be beneficial for the paper to conduct an:

             - using the sine heuristic vs. standard Euclidean ,

            - the benefit of B-spline smoothing (with vs. without),

            - the contribution of each term in the improved DWA cost function.

5. The introduction section could include more recent papers from 2023–2024 on combined algorithms (e.g., reinforcement learning + DWA) as a comparative focus.

Author Response

Comments 1：In Eq. (3), sin(α/β) is introduced without defining α and β clearly in mathematical terms (are they scalars or vectors, and are α and β angles or parameters?).

And why is chosen sin function heuristics?

Response 1：Thank you very much for your valuable suggestion. As you pointed out, the parameters α and β were not clearly defined in mathematical terms in the original manuscript. Following your recommendation, we have revised the description in the text (Lines 182–187) to clearly specify that both α and β are scalar Euclidean distances rather than angular quantities. Furthermore, we have added an explanation to justify the choice of the sine function in the heuristic design, highlighting its role in enabling a smooth and nonlinear transition from global exploration to local refinement.

 

Comments 2：While for JPS metrics (time, number of nodes) are shown, for DWA the section is more qualitative. Although the text explains what is being improved, but there is no table or graphic showing the effects of the new heuristics or route prescription, for example: Reduction in number of nodes, reduction in computation time, Improvement in path length

Response 2：Thank you very much for your insightful comments. The main contributions of this work are concentrated on the improvements to the JPS algorithm, including the introduction of a quadrant-based pruning strategy guided by the target direction, as well as a dynamically sine-weighted heuristic function—both serving as the core innovations in the global planning phase. In contrast, the enhancements to the DWA module are relatively limited, primarily involving the addition of a target orientation factor, a safety distance penalty, and a normalization scheme within the existing cost function framework.

Given this emphasis, we chose to qualitatively demonstrate the improvements to the DWA component through the simulation results and velocity response curves presented in Figures 8–10, which effectively reflect its real-time responsiveness and dynamic obstacle avoidance capability. Therefore, we did not provide separate quantitative tables for each term in this section. A more detailed evaluation of the modified DWA cost terms and computational performance will be included in future work to strengthen the analysis of its effectiveness and real-time applicability.

 

Comments 3：No computational cost analysis The algorithm is for real-time use, but it is not stated how fast it runs, think to add time analysis (fps or time per iteration)

Response 3：Thank you very much for your valuable suggestion. As per your advice, we have included a runtime analysis to evaluate the real-time performance of the proposed hybrid algorithm. Specifically, we conducted 10 simulation runs in both the standard and complex environments and reported the average total runtime along with the estimated per-step computational cost, based on the number of trajectory points. This new content has been added in Lines 461–469 of the revised manuscript.

 

Comments 4：It would be beneficial for the paper to conduct an:

             - using the sine heuristic vs. standard Euclidean ,

            - the benefit of B-spline smoothing (with vs. without),

            - the contribution of each term in the improved DWA cost function.

Response 4：Thank you for your constructive suggestion. As demonstrated in Figures 5, 6, and 7, the simulations already include comparative analyses on: (1) the search performance differences between the sine-based heuristic and the traditional Euclidean heuristic, and (2) the trajectory quality with and without B-spline smoothing. Regarding the contribution of each term in the improved DWA cost function, we have already provided a detailed explanation in the response to Comment 2.

 

Comments 5：The introduction section could include more recent papers from 2023–2024 on combined algorithms (e.g., reinforcement learning + DWA) as a comparative focus.

Response 5：Thank you for your helpful suggestion. We have made the corresponding revision in the introduction section as requested. (Lines 65–81)

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript proposes a hybrid path planning algorithm for mobile robots that combines an improved Jump Point Search (JPS) for global planning with an enhanced Dynamic Window Approach (DWA) for local planning. In the global planning phase, the method introduces a quadrant-based pruning strategy guided by the target direction and a sine-based heuristic function to reduce the search space and accelerate computation. Additionally, natural jump points are preserved to ensure path continuity, and cubic B-spline interpolation is used to smooth the global path. In the local planning phase, the traditional DWA is modified by incorporating a target orientation factor, a safety distance penalty term, a normalisation mechanism for the cost function, and a dynamic weighting strategy that adapts based on the proximity of obstacles. The proposed method is evaluated through simulation experiments in both static and dynamic environments, including those with unknown and moving obstacles. According to the abstract, the improved JPS reduces search time by 36.7 percent and node expansions by 60.9 percent compared to the standard approach, with similar path lengths. When integrated with the enhanced DWA, the robot is able to adapt effectively to changing environments and ensure safe and efficient navigation. The strengths of the manuscript lie in its well-structured integration of global and local planning methods, meaningful performance improvements, and attention to real-world challenges such as obstacle avoidance and path smoothness. The inclusion of B-spline smoothing and safety-aware local planning addresses practical deployment needs. The simulation results suggest strong performance gains, and the system appears robust to various environmental conditions. However, the paper also presents some limitations. First, the novelty of the work appears to be incremental; the core components, JPS, DWA, B-spline smoothing, and cost-function tuning, are all established techniques. The contribution may lie more in engineering integration than in novel algorithmic development. Second, the abstract and conclusion lack details about the evaluation setup, such as map types, number of trials, obstacle densities, and statistical significance of the results. This limits the ability to assess the rigour and reproducibility of the claims. Third, while the method performs well in simulation, there is no mention of real-world experiments; these are only proposed as future work. This gap raises concerns about the practical applicability of the algorithm outside controlled environments. Finally, some of the terminology used, such as "sine-enhanced heuristic" and "natural jump points," could be better defined or clarified for the reader.

Author Response

Comments ：The manuscript proposes a hybrid path planning algorithm for mobile robots that combines an improved Jump Point Search (JPS) for global planning with an enhanced Dynamic Window Approach (DWA) for local planning. In the global planning phase, the method introduces a quadrant-based pruning strategy guided by the target direction and a sine-based heuristic function to reduce the search space and accelerate computation. Additionally, natural jump points are preserved to ensure path continuity, and cubic B-spline interpolation is used to smooth the global path. In the local planning phase, the traditional DWA is modified by incorporating a target orientation factor, a safety distance penalty term, a normalisation mechanism for the cost function, and a dynamic weighting strategy that adapts based on the proximity of obstacles. The proposed method is evaluated through simulation experiments in both static and dynamic environments, including those with unknown and moving obstacles. According to the abstract, the improved JPS reduces search time by 36.7 percent and node expansions by 60.9 percent compared to the standard approach, with similar path lengths. When integrated with the enhanced DWA, the robot is able to adapt effectively to changing environments and ensure safe and efficient navigation. The strengths of the manuscript lie in its well-structured integration of global and local planning methods, meaningful performance improvements, and attention to real-world challenges such as obstacle avoidance and path smoothness. The inclusion of B-spline smoothing and safety-aware local planning addresses practical deployment needs. The simulation results suggest strong performance gains, and the system appears robust to various environmental conditions. However, the paper also presents some limitations. First, the novelty of the work appears to be incremental; the core components, JPS, DWA, B-spline smoothing, and cost-function tuning, are all established techniques. The contribution may lie more in engineering integration than in novel algorithmic development. Second, the abstract and conclusion lack details about the evaluation setup, such as map types, number of trials, obstacle densities, and statistical significance of the results. This limits the ability to assess the rigour and reproducibility of the claims. Third, while the method performs well in simulation, there is no mention of real-world experiments; these are only proposed as future work. This gap raises concerns about the practical applicability of the algorithm outside controlled environments. Finally, some of the terminology used, such as "sine-enhanced heuristic" and "natural jump points," could be better defined or clarified for the reader.

 

Response to Comment 1:
Thank you for your valuable suggestion. The proposed sine-enhanced heuristic, quadrant-based pruning strategy, and dynamically weighted DWA are, to the best of our knowledge, integrated for the first time within a unified global–local path planning framework. This integration enables each component to complement the others effectively, resulting in significantly better performance compared with baseline methods, as demonstrated in our simulation results. I have made the requested revision by adding a paragraph outlining the main contributions after the introduction section. (Lines 94–104)

Response to Comment 2:
Thank you for pointing this out. Due to the word limit in the abstract, we only briefly mentioned the most important results. Detailed descriptions of the experimental setup, including map types, number of trials, and obstacle configurations, are provided in the experimental analysis section of the manuscript. We have further supplemented this section to enhance the clarity and completeness of the evaluation context.

Response to Comment 3:
Thank you for your comment. The absence of real-world experiments in the current study is mainly due to hardware limitations. We fully recognize the importance of validating the proposed method in practical scenarios, and implementing real-world tests will be a priority in our future work.

Response to Comment 4:
Thank you for noting the lack of clarity in certain terminology. We have revised the manuscript to provide clear definitions for "sine-enhanced heuristic" and "natural jump points".

Reviewer 3 Report

Comments and Suggestions for Authors

Authors in this work present a path planning algorithm for mobile robots based on a hybrid jump point search and dynamic window approach. In general, the work is adequately developed. The authors should consider expanding the experimental design to address more cases. Some aspects to consider to improve the paper are:

1. Since the proposal addresses various aspects, it is recommended to clarify the main contribution made in the paper.

2. In the final part of the introduction, it is recommended to include the organization of the document, correctly citing the respective sections.

3. For continuity in reading, it is recommended to include an introductory paragraph in Section 2.

4. The caption for Figure 1 is not located on the same page as the figure. This aspect must be reviewed and corrected.

5. Being a fundamental part of the paper, it is recommended to expand the description of Figure 4 and Algorithm 1 (page 10).

6. It is important to improve the quality of Figure 10. It is also recommended to expand the description of this figure.

7. It is recommended to consider expanding the experimental design to address a broader range of cases.

8. Considering the limitations and other research directions, it is recommended to include a discussion section that addresses these aspects.

9. Could the algorithm be improved by incorporating a computational intelligence technique, such as fuzzy logic?

Author Response

Comments 1：Since the proposal addresses various aspects, it is recommended to clarify the main contribution made in the paper.

Response 1：Thank you for your suggestion. I have made the requested revision by adding a paragraph outlining the main contributions after the introduction section. (Lines 94–104)

 

Comments 2： In the final part of the introduction, it is recommended to include the organization of the document, correctly citing the respective sections.

Response 2：Thank you for your suggestion. I have made the corresponding revision as requested and added the article structure description at the end of the introduction section. (Lines 105–109)

 

Comments 3：For continuity in reading, it is recommended to include an introductory paragraph in Section 2.

Response 3：Thank you for your valuable suggestion. I have made the corresponding revision as requested by adding an introductory paragraph at the beginning of Section 2 to improve the continuity of the reading. (Lines 111–119)

 

Comments 4： The caption for Figure 1 is not located on the same page as the figure. This aspect must be reviewed and corrected.

Response 4：Thank you for your comment. In the submitted manuscript, the caption for Figure 1 is already located on the same page as the figure. I have rechecked the layout to ensure this is correct.

 

Comments 5：Being a fundamental part of the paper, it is recommended to expand the description of Figure 4 and Algorithm 1 (page 10).

Response 5：Thank you for your valuable suggestion. Figure 4 (flowchart of the hybrid algorithm) and Algorithm 1 (pseudocode) have been carefully designed to present the workflow and logic of the proposed method in a concise yet comprehensive manner. We believe the current level of detail is sufficient for readers to follow the methodology without ambiguity. Nevertheless, we appreciate your comment and will consider providing additional elaboration in future revisions should it be deemed necessary.

 

Comments 6：It is important to improve the quality of Figure 10. It is also recommended to expand the description of this figure.

Response 6：We appreciate your suggestion. In the revised manuscript, the quality of Figure 10 has been improved to enhance visual clarity, and the description of this figure has been expanded to provide a more detailed explanation of the presented results. (Lines 475–485)

 

Comments 7：It is recommended to consider expanding the experimental design to address a broader range of cases.

Response 7：We appreciate your suggestion. The current experimental design was carefully formulated to align with the primary objectives and scope of this study, and it includes a variety of representative scenarios to validate the proposed method. While expanding to a broader range of cases is beyond the present scope, we agree that this is a valuable direction for future research and will consider it in our subsequent work.

 

Comments 8：Considering the limitations and other research directions, it is recommended to include a discussion section that addresses these aspects.

Response 8：Thank you for your valuable suggestion. As per your recommendation, we have incorporated the discussion of the study’s limitations and future research directions into the conclusion section of the revised manuscript. (Lines 513–518)

 

Comments 9：Could the algorithm be improved by incorporating a computational intelligence technique, such as fuzzy logic?

Response 9：Thank you for your insightful suggestions. Incorporating computational intelligence techniques such as fuzzy logic could indeed further enhance the adaptability and decision-making capabilities of the proposed algorithm, especially in highly uncertain or unstructured environments. However, this study focused on integrating and optimizing the JPS and DWA algorithms within a unified framework, so such an extension is beyond the current scope. We agree that this is a promising research direction, and we plan to investigate the integration of fuzzy logic or other computational intelligence methods in future work to further improve the algorithm's performance.

 

We have professionally proofread the entire manuscript to improve English clarity and readability. The proofreading certificate for professional editing services is attached for your reference.

Author Response File: Author Response.pdf