A Multi-Active and Multi-Passive Sensor Fusion Algorithm for Multi-Target Tracking in Dense Group Clutter Environments
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsFor multi-target tracking under the MAMPS with dense group clutter, the clutter pre-processing algorithm of CFDP with double threshold screening included and the measurement initial screening algorithm based on EKF are proposed to solve those problems that clutter and high computational complexity due to an explosion in measurement association groups’ number. Simultaneously, the MAMPS-MAF-DNN algorithm is proposed to solve the problem of ineffective fusion of multi-source sensor information. The algorithm proposed in the paper has a certain degree of innovation and is well written, which can highlight the authors’ outstanding work. However, before considering publication, there are still a series of issues that urgently need improvement, as follows:
1. In Abstract, suggest modifying “Due to the differences of measurement information characteristics in MAMPS fusion” to “Due to the difference of measurement information characteristics in MAMPS fusion”.
2. In Introduction, since the abbreviation of the probability hypothesis density has already appeared, i.e., PHD, it is recommended to modify “cardinalized probability hypothesis density (CPHD)” to “cardinalized PHD (CPHD)”.
3. After reading the entire paper, I found that this work is an extension of the conference paper [18]. Suggest the authors to emphasize the innovation of the proposed algorithm in the introduction.
4. In equation (2), the authors did not provide a standardized definition for scalar “Omega”, lacking time k. It is suggested to revise it.
5. In Figure 7, the coordinates of point P are incorrect. It is recommended to modify them to the correct coordinates.
6. Considering the completeness of the conclusion, it is recommended to include future prospects in Section 7.
In the second paragraph of Section 1, for ease of understanding, “the above-mentioned multi-target tracking algorithms” is suggested to modify to “the multi-target tracking algorithms mentioned above”.
Author Response
Comment 1: In Abstract, suggest modifying “Due to the differences of measurement information characteristics in MAMPS fusion” to “Due to the difference of measurement information characteristics in MAMPS fusion”.
Response 1: First of all, thank you for your important revised suggestions. According to your suggestion, we have revised this sentence in Abstract.
Comment 2: In Introduction, since the abbreviation of the probability hypothesis density has already appeared, i.e., PHD, it is recommended to modify “cardinalized probability hypothesis density (CPHD)” to “cardinalized PHD (CPHD)”.
Response: First of all, thank you for your kind reminder! According to your suggestion, we have modified “cardinalized probability hypothesis density (CPHD)” to “cardinalized PHD (CPHD)” in the first paragraph of Section 1.
Comment 3: After reading the entire paper, I found that this work is an extension of the conference paper [18]. Suggest the authors to emphasize the innovation of the proposed algorithm in the introduction.
Response: First of all, thank you for your important revised suggestions. I agree that this paper is an extension of their previous work [18]. According to your suggestion, in order to emphasize the innovation of the proposed algorithm in this work, we have added some content to explain the specific differences between this work and the conference paper in the fifth paragraph of Section 1.
Comment 4: In equation (2), the authors did not provide a standardized definition for scalar “Omega”, lacking time k. It is suggested to revise it.
Response: First of all, thank you for your kind reminder! According to your suggestion, we have corrected this mistake in equation (2).
Comment 5: In Figure 7, the coordinates of point P are incorrect. It is recommended to modify them to the correct coordinates.
Response: First of all, thank you for your kind reminder! According to your suggestion, we have corrected this mistake in Figure 7.
Comment 6: Considering the completeness of the conclusion, it is recommended to include future prospects in Section 7.
Response: First of all, thank you for your important revised suggestions. According to your suggestion, we have added future prospects in the last paragraph of Section 7.
Finally, we deeply appreciate the constructive and valuable comments and suggestions from you, which can be of great help to improve the quality of our manuscript.
Comments on the Quality of English Language: In the second paragraph of Section 1, for ease of understanding, “the above-mentioned multi-target tracking algorithms” is suggested to modify to “the multi-target tracking algorithms mentioned above”.
Response: First of all, thank you for your important revised suggestions. According to your suggestion, we have modified “the above-mentioned multi-target tracking algorithms” to “the multi-target tracking algorithms mentioned above” in the second paragraph of Section 1. In addition, we have further polished the language throughout the manuscript to increase its readability.
Reviewer 2 Report
Comments and Suggestions for AuthorsFor the measurement fusion problem of different types of sensors, the paper proposes a novel MAMPS fusion algorithm, which is based on the centralized measurement association fusion (MAF) and distributed deep neural networks (DNN) track fusion, named the MAMPS-MAF-DNN algorithm. In the proposed algorithm, firstly, to reduce the impact of the dense group clutter, a clutter pre-processing algorithm is elaborated, combining advantages of the CFDP (cluster by finding density peaks) and double threshold screening algorithms. Then, for the single-active and multi-passive sensor (SAMPS) system, a centralized MAF algorithm based on angle information is developed. Finally, the SAMPS-MAF algorithm is extended to the MAMPS system within the DNN framework. The proposed algorithm has good innovation and is an excellent paper. However, the following issues need to be improved before considering publication. The detailed problems are shown as follows.
1. Multi-target tracking appears multiple times in this paper. In order to express it more concisely, it is recommended to use its abbreviation, i.e., MTT, to represent the full name of multi-target tracking.
2. In Section 1, to highlight the referential relationship, it is suggested to modify “However, as the number of sensors and measurements increases, the computational complexity increases rapidly” as “However, as the number of sensors and measurements increases, its computational complexity increases rapidly”.
3. This paper lacks a description of the overall framework, so it is suggested to add it at the end of Section 1.
4. The first and second paragraphs of Section 3 are relatively short, it is suggested to merge them into one paragraph.
5. After some equations, considering rigor, it is recommended to remove the comma after “where”, such as the comma after “where” after equations (7) and (8).
6. This paper is an extension of their previous work [18]. Suggest the authors to add specific differences between the proposed algorithm in this work and previous conference paper.
Comments on the Quality of English LanguageMinor editing of English language required.
Author Response
Comment 1: Multi-target tracking appears multiple times in this paper. In order to express it more concisely, it is recommended to use its abbreviation, i.e., MTT, to represent the full name of multi-target tracking.
Response 1: First of all, thank you for your important revised suggestions. According to your suggestion, we have used MTT to represent the full name of multi-target tracking in the entire manuscript.
Comment 2: In Section 1, to highlight the referential relationship, it is suggested to modify “However, as the number of sensors and measurements increases, the computational complexity increases rapidly” as “However, as the number of sensors and measurements increases, its computational complexity increases rapidly”.
Response: First of all, thank you for your kind reminder! According to your suggestion, we have revised this sentence to highlight the referential relationship in the third paragraph of Section 1.
Comment 3: This paper lacks a description of the overall framework, so it is suggested to add it at the end of Section 1.
Response: First of all, thank you for your important revised suggestions. We have added a description of the overall framework in the last paragraph of Section 1 according to your suggestion.
Comment 4: The first and second paragraphs of Section 3 are relatively short, it is suggested to merge them into one paragraph.
Response: First of all, thank you for your important revised suggestions. According to your suggestion, we have merged these two paragraphs into one paragraph, i.e., the first paragraph of Section 3.
Comment 5: After some equations, considering rigor, it is recommended to remove the comma after “where”, such as the comma after “where” after equations (7) and (8).
Response: First of all, thank you for your kind reminder! In the entire manuscript, we have removed the comma after “where” which appears after equations.
Comment 6: This paper is an extension of their previous work [18]. Suggest the authors to add specific differences between the proposed algorithm in this work and previous conference paper.
Response: First of all, thank you for your important revised suggestions. I agree that this paper is an extension of their previous work [18]. According to your suggestion, we have added some content to explain the specific differences between this work and the conference paper in the fifth paragraph of Section 1.
Finally, we deeply appreciate the constructive and valuable comments and suggestions from you, which can be of great help to improve the quality of our manuscript.