Real-Time Data Acquisition System for Array MIMU Based on FPGA+ARM

3. Point-by-point response to Comments and Suggestions for Authors

Comments 1: The authors present a real-time FPGA+ARM-based data acquisition system for MIMU arrays that is capable of simultaneous reading of 60 gyroscopes, which significantly improves data accuracy and stability compared to currently known technologies that utilize sequential reading of multiple sensors.

This paper is well-organized. It details the overall design scheme for the array MIMU real-time data acquisition system, covering both hardware circuit design and software code development. It also elaborates on the quaternion-based attitude update method and position update method.

The proposed data acquisition system was analyzed and validated through static experiments and practical pipeline scenario testing. The experiment results show that the system substantially improves gyroscope accuracy, highlighting the array design’s effectiveness in mitigating random errors and enabling high-precision, real-time measurements.

Response 1: Thank you very much for your positive evaluation and affirmation of this paper! Your recognition is a tremendous encouragement for our research work.

4. Response to Comments on the Quality of English Language

Point 1:The English could be improved to more clearly express the research.

Response 1:We sincerely appreciate your valuable feedback regarding the English expression in our paper. We fully agree that precise and clear English presentation is crucial for readers. We have systematically refined and improved the entire text, focusing on enhancing the clarity of research content presentation, logical coherence, and accuracy in terminology usage. All relevant sections have been highlighted in blue font within the paper.

Comments 1: Although MIMU (Micro-Inertial Measurement Unit) and MEMS (Micro-Electro-Mechanical Systems) are defined in the abstract—an independent preface module separate from the main text—the introduction directly uses these abbreviations without re-stating their full names upon first occurrence. This may confuse readers, please make corresponding annotations.

Response 1:We sincerely appreciate your meticulous review comments. We fully agree with your suggestions and have completed the corresponding revisions accordingly. Please refer to the first paragraph of the introduction in the revised manuscript for the specific changes. Once again, thank you for helping us enhance the standardization of our paper.

Comments 2: Figure 2 (Topology) and Figure 4 (Program execution flow) are somewhat generic. It is recommended to add timing diagrams for the FPGA internal SPI controller to demonstrate how "synchronization" is achieved at the nanosecond/microsecond level.

Response 2: Thank you very much for your valuable feedback. We have carefully incorporated your suggestions by adding the timing diagram for the FPGA's internal SPI controller along with corresponding explanations to ensure the system can synchronously read data from 60 sensors. The modified content is highlighted in blue font in the fourth paragraph of Section 2.2 in Chapter 2.

Comments 3: There is inconsistency in the time representation of the horizontal axis across figures: Figure 7 uses "Time (s)", while Figures 9 and 10 adopt "time/s". Please standardize and ensure consistency in the format of the paper.

Response 3: Thank you very much for your meticulous and professional review. Your valuable feedback regarding inconsistent image formats has been noted. We have thoroughly checked and revised the manuscript, ensuring uniformity in the horizontal axis labels for Figures 9 and 10. Once again, we sincerely appreciate your review.

Comments 4: The font size of the charts lacks consistency: some of the legend texts in the charts (such as Figure 13) are too small to read clearly. Please adjust the clarity of the charts in the paper.

Response 4: We sincerely appreciate your valuable feedback. We fully agree with your observations regarding the small font size and insufficient clarity in the figures, which indeed hindered readers' comprehension of key data findings. As the most intuitive medium for presenting research outcomes, the clarity and readability of figures are paramount. Following your suggestions, we have systematically reviewed and comprehensively optimized all figures throughout the manuscript, including Figure 13 which you specifically highlighted. We are deeply grateful for your assistance in enhancing the presentation quality of our paper.

Comments 5: In the pipeline test (Section 4.2), the authors report positioning errors (horizontal < 0.0774m, elevation < 0.0351m). The text mentions comparing against "true position coordinates" provided by a "collaborating organisation.” It is mandatory to specify how this "Ground Truth" was obtained. Additionally, pipelines often have magnetic interference; since the paper mentions using a Magnetometer, please explain how magnetic interference is rejected or handled in the ferromagnetic pipeline environment.

Response 5: We sincerely appreciate your valuable feedback. The points you raised regarding “methods for acquiring actual ground data” and “handling magnetic interference in ferromagnetic pipeline environments” are indeed critical issues that must be explicitly addressed in this experimental validation. We fully concur with your perspective. Below are detailed explanations for each issue: First, the actual pipeline location coordinates were measured by our collaborating institution using millimeter-grade total stations and electronic levels, in accordance with the Hebei Provincial Local Metrology Technical Specifications. The relevant modifications have been indicated in blue font in the third paragraph of Section 2, Chapter 4 of the paper. Second, regarding magnetic interference within the pipeline, we not only selected a high-precision magnetometer but also calibrated it using multi-position ellipsoid fitting, significantly reducing this impact. Furthermore, during pipeline inspection, we relied entirely on the magnetometer to determine the initial heading angle of the carrier. To further ensure data accuracy, multiple sets of magnetic field readings were collected during the magnetic data acquisition phase. The program automatically filters out outliers and averages the remaining magnetic data to guarantee the precision of the initial heading angle. The relevant modifications have been highlighted in blue in the first paragraph of Section 2, Chapter 4 of the paper. We sincerely appreciate your valuable feedback, which has provided us with an opportunity to present the technical details and innovations of this system more comprehensively.

Comments 6: The future outlook is overly vague: it does not specify the integration pathway of the proposed algorithm with existing array fusion technologies, nor propose specific directions such as expanding application scenarios or optimizing sensor layout. Please provide guidance on the continuity of the research to reflect its depth and scalability.

Response 6: Thank you very much for your valuable feedback. Your observation that the “future outlook is too vague” accurately highlights a shortcoming in our initial draft. Based on your suggestion, we have refined and elaborated the “Future Outlook” section to more clearly articulate the continuity, depth, and scalability of our research. These additions have been incorporated into the conclusion of Chapter 5 and are indicated in blue font. The specific additions are as follows: Subsequent work will prioritize the integration of real-time noise reduction algorithms based on Kalman filtering while strictly maintaining the system's real-time performance. Specifically, this involves constructing a two-stage noise reduction architecture combining “array fusion + Kalman filtering”—using the hardware-level mean fusion from 60 MIMU channels as the observation input for Kalman filtering, and employing the filtering algorithm to dynamically estimate gyro random errors for real-time compensation. Next, optimize the adaptive filtering parameter adjustment strategy. Finally, embed this integrated algorithm into an FPGA to achieve a secondary enhancement in gyroscope measurement accuracy, further suppressing inherent gyroscope random errors. Simultaneously, expand its application scenarios to high-precision domains such as drones, robotics, and underground space exploration.Once again, we sincerely appreciate your thorough review and constructive input.

4. Response to Comments on the Quality of English Language

Point 1:The English is fine and does not require any improvement.

Response 1: Thank you very much for your affirmation of the quality of our paper's English expression!