pith. sign in

arxiv: 1907.03305 · v1 · pith:6WS5K5LTnew · submitted 2019-07-07 · 💻 cs.RO · cs.SY· eess.SY

System Architecture for Real-time Surface Inspection Using Multiple UAVs

Van Truong Hoang , Manh Duong Phung , Tran Hiep Dinh , Quang P. Ha This is my paper

Pith reviewed 2026-05-25 01:24 UTC · model grok-4.3

classification 💻 cs.RO cs.SYeess.SY
keywords UAVsurface inspectionparticle swarm optimizationIoTimage processingdefect detectionmulti-UAV formationreal-time control
0
0 comments X

The pith

Multiple UAVs inspect surfaces in real time by using angle-encoded particle swarm optimization to plan and assign paths, then transmit data over an IoT network for histogram-based defect detection.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper describes a coordinated multi-UAV system for surface inspection where an angle-encoded particle swarm optimization algorithm generates inspection paths and redistributes them among the vehicles. Communication occurs through IoT boards that enable real-time data transfer to remote units. An online histogram-based image processing technique then identifies potential defects during flight. Simulations and experiments are used to check that the combined path planning, networking, and detection steps operate together without breakdown.

Core claim

The architecture coordinates UAVs into a formation by applying angle-encoded particle swarm optimisation to create and allocate inspecting paths, equips the vehicles with IoT boards for network and processing functions, streams collected data in real time to remote computers, and applies a histogram method for online detection of surface damage or defects.

What carries the argument

Angle-encoded particle swarm optimisation that generates inspecting paths and redistributes them to each UAV, integrated with IoT communication links and a histogram-based online image processing technique.

If this is right

  • Data from multiple UAVs reaches remote units in real time for immediate analysis.
  • Defect detection occurs online without requiring post-flight processing.
  • The same path-planning step can be reused for different surface shapes by redistributing the route among available vehicles.
  • The IoT layer supports simultaneous network access for all vehicles in the formation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The approach could be tested on non-planar or moving objects if the optimisation step is rerun at regular intervals.
  • Replacing the histogram step with other simple metrics might reveal whether the real-time constraint is the main limit on detection accuracy.
  • The architecture might support inspection of infrastructure such as bridges or pipelines once the path generator accounts for vertical surfaces.

Load-bearing premise

The paths produced by angle-encoded particle swarm optimisation remain collision-free and feasible once IoT networking and real-time histogram processing are added, and the histograms can separate defects from normal surface changes under actual flight lighting and motion.

What would settle it

Run the generated paths in a physical test flight and check whether any UAVs collide or whether the histogram detector misses known defects or flags normal variations as damage.

Figures

Figures reproduced from arXiv: 1907.03305 by Manh Duong Phung, Quang P. Ha, Tran Hiep Dinh, Van Truong Hoang.

Figure 1
Figure 1. Figure 1: Data communication structure B. Communication protocols Along with the hardware, transport protocols play an im￾portant role in ensuring the security and efficiency of the data exchanged. The most popular transport protocols for IoT include the Transmission Control Protocol (TCP), User Datagram Protocol (UDP), and Real-Time Transport Protocol (RTP). While TCP was originally designed for the reliable transm… view at source ↗
Figure 2
Figure 2. Figure 2: System architecture for multi-UAV surface inspection [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Inertial and formation frames in UAV formation [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Mission Planner incorporating Google Satellite Map to [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Pseudo-code for path generation process based on (10) to maintain the shape. During the flight, on￾board computers calculate the inverse kinematics, obtain posi￾tion errors with respect to their neighbours and the formation centroid, and then drive these errors to zero with the tracking control. V. SURFACE INSPECTION For defect detection, images of the inspected surface taken by UAVs are sent to RCU. Due t… view at source ↗
Figure 6
Figure 6. Figure 6: Flowchart of the defect detection algorithm [PITH_FULL_IMAGE:figures/full_fig_p007_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Pseudo code for defect detection VI. RESULTS The performance of our proposed system has been evaluated in a number of surface inspection tasks. This section describes the system testbed and experimental results. A. Experimental setup The UAVs used in this study is the 3DR Solo drone shown in [PITH_FULL_IMAGE:figures/full_fig_p007_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: The 3DR Solo testbed [PITH_FULL_IMAGE:figures/full_fig_p007_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Convergence comparison between conventional PSO [PITH_FULL_IMAGE:figures/full_fig_p008_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Bridge inspection with UAV formation (TLBO) [42] [PITH_FULL_IMAGE:figures/full_fig_p008_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Trajectories of three UAVs tracking the planned paths [PITH_FULL_IMAGE:figures/full_fig_p009_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Altitudes of the three UAVs in the formation test [PITH_FULL_IMAGE:figures/full_fig_p009_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Errors between the planned and flown paths [PITH_FULL_IMAGE:figures/full_fig_p009_13.png] view at source ↗
Figure 15
Figure 15. Figure 15: Defect detection results. First row: image name, original image, ground truth, our result, Sauvola; second row: detection [PITH_FULL_IMAGE:figures/full_fig_p010_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: Defect detection results. First row: image name, original image, ground truth, our result, Sauvola; second row: detection [PITH_FULL_IMAGE:figures/full_fig_p011_16.png] view at source ↗
read the original abstract

This paper presents a real-time control system for surface inspection using multiple unmanned aerial vehicles (UAVs). The UAVs are coordinated in a specific formation to collect data of the inspecting objects. The communication platform for data transmission is based on the Internet of Things (IoT). In the proposed architecture, the UAV formation is established via using the angle-encoded particle swarm optimisation to generate an inspecting path and redistribute it to each UAV where communication links are embedded with an IoT board for network and data processing capabilities. Data collected are transmitted in real time through the network to remote computational units. To detect potential damage or defects, an online image processing technique is proposed and implemented based on histograms. Extensive simulation, experiments and comparisons have been conducted to verify the validity and performance of the proposed system.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 0 minor

Summary. The manuscript presents a system architecture for real-time surface inspection with multiple UAVs. UAV formation and inspection paths are generated and redistributed using angle-encoded particle swarm optimization; communication and data processing use IoT boards; data are transmitted in real time to remote units; and defect detection employs an online histogram-based image processing technique. The authors state that extensive simulations, experiments, and comparisons verify the validity and performance of the proposed system.

Significance. If the experimental validation holds, the work could provide a practical integrated architecture combining PSO-based path planning, IoT-enabled real-time networking, and histogram-based defect detection for multi-UAV surface inspection tasks.

major comments (3)
  1. [Abstract] Abstract: the central claim of real-time operation and verified performance rests on the statement that 'extensive simulation, experiments and comparisons have been conducted,' yet the abstract (and by extension the manuscript) provides no quantitative metrics, error bars, number of trials, or baseline comparisons to support this.
  2. The histogram-based online image processing for defect detection is presented without any evaluation of robustness to UAV-specific conditions such as illumination gradients, specular reflections, motion blur, or viewpoint changes; global intensity histograms are known to be sensitive to these factors, and no demonstration is given that the chosen features remain discriminative under actual flight envelopes.
  3. The integration claim—that angle-encoded PSO paths remain collision-free and feasible once IoT network latency and real-time image processing are added—is asserted without any analysis or results showing that the added components preserve the detection margin or path validity.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment below and will revise the manuscript accordingly where appropriate to strengthen the presentation of results and clarify limitations.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim of real-time operation and verified performance rests on the statement that 'extensive simulation, experiments and comparisons have been conducted,' yet the abstract (and by extension the manuscript) provides no quantitative metrics, error bars, number of trials, or baseline comparisons to support this.

    Authors: We agree that the abstract would benefit from explicit quantitative metrics to support the claims. The body of the manuscript contains results from simulations and experiments, including performance comparisons. In the revised manuscript, we will update the abstract to summarize key metrics such as inspection times, detection rates, and baseline comparisons drawn from the experimental sections. revision: yes

  2. Referee: The histogram-based online image processing for defect detection is presented without any evaluation of robustness to UAV-specific conditions such as illumination gradients, specular reflections, motion blur, or viewpoint changes; global intensity histograms are known to be sensitive to these factors, and no demonstration is given that the chosen features remain discriminative under actual flight envelopes.

    Authors: The histogram method was selected primarily for its computational simplicity to support real-time IoT-based processing. Experiments were conducted under the lighting and motion conditions of our test setups. We acknowledge that global histograms can be sensitive to the listed factors and that dedicated robustness tests under varied flight conditions are not included. We will add a limitations paragraph in the revised manuscript discussing these sensitivities and identifying them as directions for future enhancement. revision: partial

  3. Referee: The integration claim—that angle-encoded PSO paths remain collision-free and feasible once IoT network latency and real-time image processing are added—is asserted without any analysis or results showing that the added components preserve the detection margin or path validity.

    Authors: The reported experiments incorporate simultaneous path execution, IoT data transmission, and online image processing. However, we did not include a dedicated analysis isolating the effects of network latency on path feasibility. We will add a short analysis subsection in the revised version that reports observed latencies from the experiments and discusses their influence on overall system timing and path validity. revision: yes

Circularity Check

0 steps flagged

No circularity: architecture description is self-contained

full rationale

The paper describes an integrated UAV inspection system that combines angle-encoded PSO for path generation, IoT-based communication, and histogram-based image processing, with validation via simulation and experiments. No equations, predictions, or derivations are presented that reduce a claimed output to a fitted input or self-citation by construction. The central claims rest on implementation choices and empirical verification rather than any closed mathematical chain internal to the paper itself.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The paper is an applied systems description. No explicit free parameters, mathematical axioms, or newly postulated physical entities are introduced beyond standard algorithmic tuning constants that are not detailed in the abstract.

pith-pipeline@v0.9.0 · 5675 in / 1190 out tokens · 22391 ms · 2026-05-25T01:24:58.412704+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

43 extracted references

  1. [1]

    Unmanned aircraft system applications in construction,

    M. C. Tatum and J. Liu, “Unmanned aircraft system applications in construction,” Procedia Engineering, vol. 196, pp. 167–175, 2017

    2017

  2. [2]

    Automatic detection of wind turbine blade surface cracks based on uav-taken images,

    L. Wang and Z. Zhang, “Automatic detection of wind turbine blade surface cracks based on uav-taken images,” IEEE Transactions on Industrial Electronics, vol. 64, no. 9, pp. 7293–7303, 2017

    2017

  3. [3]

    Automatic power line inspection using uav images,

    Y . Zhang, X. Yuan, t. Li, and S. Chen, “Automatic power line inspection using uav images,” Remote Sensing , vol. 9, no. 8, p. 824, 2017

    2017

  4. [4]

    Quality assessment of unmanned aerial vehicle (uav) based visual inspection of structures,

    G. Morgenthal and N. Hallermann, “Quality assessment of unmanned aerial vehicle (uav) based visual inspection of structures,” Advances in Structural Engineering , vol. 17, no. 3, pp. 289–302, 2014

    2014

  5. [5]

    A self organizing map optimization based image recognition and processing model for bridge crack inspection,

    J.-H. Chen, M.-C. Su, R. Cao, S.-C. Hsu, and J.-C. Lu, “A self organizing map optimization based image recognition and processing model for bridge crack inspection,” Automation in Construction , vol. 73, pp. 58– 66, 2017

    2017

  6. [6]

    L. J. S ´anchez-Aparicio, B. Riveiro, D. Gonzalez-Aguilera, and L. F. Ramos, “The combination of geomatic approaches and operational modal analysis to improve calibration of finite element models: A case of study in saint torcato church (guimar˜aes, portugal),” Construction and Building Materials , vol. 70, pp. 118–129, 2014

    2014

  7. [7]

    A survey of motion planning algorithms from the perspective of autonomous uav guidance,

    C. Goerzen, Z. Kong, and B. Mettler, “A survey of motion planning algorithms from the perspective of autonomous uav guidance,” Journal of Intelligent and Robotic Systems , vol. 57, no. 1, p. 65, Nov 2009

    2009

  8. [8]

    Enhanced discrete particle swarm optimization path planning for uav vision-based surface inspection,

    M. D. Phung, C. H. Quach, T. H. Dinh, and Q. Ha, “Enhanced discrete particle swarm optimization path planning for uav vision-based surface inspection,” Automation in Construction , vol. 81, pp. 25–33, 2017

    2017

  9. [9]

    Structural inspection path planning via iterative viewpoint resampling with application to aerial robotics,

    A. Bircher, K. Alexis, M. Burri, P. Oettershagen, S. Omari, T. Mantel, and R. Siegwart, “Structural inspection path planning via iterative viewpoint resampling with application to aerial robotics,” in Robotics and Automation (ICRA), 2015 IEEE International Conference on . IEEE, 2015, pp. 6423–6430

    2015

  10. [10]

    Vision and control for uavs: A survey of general methods and of inexpensive platforms for infrastructure inspection,

    K. M ´ath´e and L. Bus ¸oniu, “Vision and control for uavs: A survey of general methods and of inexpensive platforms for infrastructure inspection,” Sensors, vol. 15, no. 7, pp. 14 887–14 916, 2015

    2015

  11. [11]

    Unmanned aerial vehicles for power line inspection: A cooperative way in platforms and communications,

    C. Deng, S. Wang, Z. Huang, Z. Tan, and J. Liu, “Unmanned aerial vehicles for power line inspection: A cooperative way in platforms and communications,” J. Communications, vol. 9, no. 9, pp. 687–692, 2014

    2014

  12. [12]

    Multi-uav pre- positioning and routing for power network damage assessment,

    G. J. Lim, S. Kim, J. Cho, Y . Gong, and A. Khodaei, “Multi-uav pre- positioning and routing for power network damage assessment,” IEEE Transactions on Smart Grid , 2016

    2016

  13. [13]

    Bird flocking inspired formation control for unmanned aerial vehicles using stereo camera,

    J. Lwowski, A. Majumdar, P. Benavidez, J. J. Prevost, and M. Jamshidi, “Bird flocking inspired formation control for unmanned aerial vehicles using stereo camera,” IEEE Systems Journal , pp. 1–10, 2018

    2018

  14. [14]

    Leader-to-formation stabil- ity,

    H. G. Tanner, G. J. Pappas, and V . Kumar, “Leader-to-formation stabil- ity,” IEEE Transactions on robotics and automation , vol. 20, no. 3, pp. 443–455, 2004

    2004

  15. [15]

    Stingu and F

    E. Stingu and F. L. Lewis, A Hardware Platform for Research in Helicopter UA V Control. Dordrecht: Springer Netherlands, 2009, pp. 387–406

    2009

  16. [16]

    Coordinated trajectory planning for efficient communication relay using multiple uavs,

    S. Kim, H. Oh, J. Suk, and A. Tsourdos, “Coordinated trajectory planning for efficient communication relay using multiple uavs,”Control Engineering Practice , vol. 29, pp. 42 – 49, 2014

    2014

  17. [17]

    Continuous airborne communication relay approach using unmanned aerial vehicles,

    O. Cetin and I. Zagli, “Continuous airborne communication relay approach using unmanned aerial vehicles,” Journal of Intelligent & Robotic Systems , vol. 65, no. 1, pp. 549–562, Jan 2012

    2012

  18. [18]

    Wireless communications with unmanned aerial vehicles: opportunities and challenges,

    Y . Zeng, R. Zhang, and T. J. Lim, “Wireless communications with unmanned aerial vehicles: opportunities and challenges,” IEEE Com- munications Magazine , vol. 54, no. 5, pp. 36–42, May 2016

    2016

  19. [19]

    Vision-based monitoring of locally linear structures using an unmanned aerial vehicle,

    S. Rathinam, Z. Kim, and R. Sengupta, “Vision-based monitoring of locally linear structures using an unmanned aerial vehicle,” Journal of Infrastructure Systems, vol. 14, no. 1, pp. 52–63, 2008

    2008

  20. [20]

    A uav for bridge inspection: Visual servoing control law with orientation limits,

    N. Metni and T. Hamel, “A uav for bridge inspection: Visual servoing control law with orientation limits,” Automation in construction , vol. 17, no. 1, pp. 3–10, 2007

    2007

  21. [21]

    Towards an autonomous indoor aerial inspection vehicle,

    S. Winkvist, E. Rushforth, and K. Young, “Towards an autonomous indoor aerial inspection vehicle,” Industrial Robot: the international journal of robotics research and application , vol. 40, no. 3, pp. 196–207, 2013

    2013

  22. [22]

    Colibri: A vision-guided uav for surveillance and visual inspection,

    L. Mejias, J. F. Correa, I. Mondragon, and P. Campoy, “Colibri: A vision-guided uav for surveillance and visual inspection,” in Proceedings 2007 IEEE International Conference on Robotics and Automation , April 2007, pp. 2760–2761

    2007

  23. [23]

    Aerial robotic contact-based inspection: planning and control,

    K. Alexis, G. Darivianakis, M. Burri, and R. Siegwart, “Aerial robotic contact-based inspection: planning and control,” Autonomous Robots , vol. 40, no. 4, pp. 631–655, 2016

    2016

  24. [24]

    Uav path planning for structure inspection in windy environments,

    J. A. Guerrero and Y . Bestaoui, “Uav path planning for structure inspection in windy environments,” Journal of Intelligent & Robotic Systems, vol. 69, no. 1-4, pp. 297–311, 2013

    2013

  25. [25]

    Multicopter unmanned aerial vehicle for automated inspection of wind turbines,

    B. E. Schfer, D. Picchi, T. Engelhardt, and D. Abel, “Multicopter unmanned aerial vehicle for automated inspection of wind turbines,” in 2016 24th Mediterranean Conference on Control and Automation (MED), June 2016, pp. 244–249

    2016

  26. [26]

    Bridge deck delam- ination identification from unmanned aerial vehicle infrared imagery,

    A. Ellenberg, A. Kontsos, F. Moon, and I. Bartoli, “Bridge deck delam- ination identification from unmanned aerial vehicle infrared imagery,” Automation in Construction , vol. 72, pp. 155–165, 2016

    2016

  27. [27]

    Possibilities of use of uavs for technical inspection of buildings and constructions,

    A. Banaszek, S. Banaszek, and A. Cellmer, “Possibilities of use of uavs for technical inspection of buildings and constructions,” in IOP Conference Series: Earth and Environmental Science , vol. 95, no. 3. IOP Publishing, 2017, p. 032001

    2017

  28. [28]

    Angle-encoded swarm optimization for uav formation path planning,

    V . T. Hoang, M. D. Phung, T. H. Dinh, and Q. P. Ha, “Angle-encoded swarm optimization for uav formation path planning,” in 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , Oct 2018, pp. 5239–5244

    2018

  29. [29]

    Adaptive second- order sliding mode control of uavs for civil applications,

    V . T. Hoang, A. M. Singh, M. D. Phung, and Q. P. Ha, “Adaptive second- order sliding mode control of uavs for civil applications,” in Automation and Robotics in Construction (ISARC), 2017 International Symposium on, 2017, pp. 816–822

    2017

  30. [30]

    Sliding order and sliding accuracy in sliding mode control,

    A. Levant, “Sliding order and sliding accuracy in sliding mode control,” International journal of control , vol. 58, no. 6, pp. 1247–1263, 1993

    1993

  31. [31]

    Accelerated twisting algorithm,

    Y . Dvir and A. Levant, “Accelerated twisting algorithm,” IEEE Trans- actions on Automatic Control , vol. 60, no. 10, pp. 2803–2807, 2015

    2015

  32. [32]

    New methodolo- gies for adaptive sliding mode control,

    F. Plestan, Y . Shtessel, V . Bregeault, and A. Poznyak, “New methodolo- gies for adaptive sliding mode control,” International journal of control , vol. 83, no. 9, pp. 1907–1919, 2010

    1907

  33. [33]

    Adaptive twisting sliding mode control for quadrotor unmanned aerial vehicles,

    V . T. Hoang, M. D. Phung, and Q. P. Ha, “Adaptive twisting sliding mode control for quadrotor unmanned aerial vehicles,” in 2017 11th Asian Control Conference (ASCC) , Dec 2017, pp. 671–676

    2017

  34. [34]

    Effect of color space on color image segmentation,

    N. M. Kwok, Q. P. Ha, and G. Fang, “Effect of color space on color image segmentation,” in 2009 2nd International Congress on Image and Signal Processing, Oct 2009, pp. 1–5

    2009

  35. [35]

    A threshold selection method from gray-level histograms,

    N. Otsu, “A threshold selection method from gray-level histograms,” IEEE Trans. Syst. Man Cybern. , vol. 11, no. 285-296, pp. 23–27, 1975

    1975

  36. [36]

    Automatic thresholding for defect detection,

    H.-F. Ng, “Automatic thresholding for defect detection,” Pattern Recog- nition Letters , vol. 27, no. 14, pp. 1644 – 1649, 2006

    2006

  37. [37]

    Adaptive document image binarization,

    J. Sauvola and M. Pietikinen, “Adaptive document image binarization,” Pattern Recognition, vol. 33, no. 2, pp. 225 – 236, 2000

    2000

  38. [38]

    A new iterative triclass thresholding technique in image segmentation,

    H. Cai, Z. Yang, X. Cao, W. Xia, and X. Xu, “A new iterative triclass thresholding technique in image segmentation,” IEEE Transactions on Image Processing, vol. 23, no. 3, pp. 1038–1046, March 2014

    2014

  39. [39]

    A flexible and robust threshold selection method,

    Z. Wang, J. Xiong, Y . Yang, and H. Li, “A flexible and robust threshold selection method,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 28, no. 9, pp. 2220–2232, Sep. 2018

    2018

  40. [40]

    Dynamic measurement of computer generated image segmentations,

    M. D. Levine and A. M. Nazif, “Dynamic measurement of computer generated image segmentations,” IEEE Trans. Pattern Anal. Mach. Intell., no. 2, pp. 155–164, 1985

    1985

  41. [41]

    A new vibrational genetic algorithm enhanced with a voronoi diagram for path planning of autonomous uav,

    Y . V . Pehlivanoglu, “A new vibrational genetic algorithm enhanced with a voronoi diagram for path planning of autonomous uav,” Aerospace Science and Technology , vol. 16, no. 1, pp. 47–55, 2012

    2012

  42. [42]

    Teachinglearning-based optimiza- tion: A novel method for constrained mechanical design optimization problems,

    R. Rao, V . Savsani, and D. Vakharia, “Teachinglearning-based optimiza- tion: A novel method for constrained mechanical design optimization problems,” Computer-Aided Design, vol. 43, no. 3, pp. 303 – 315, 2011

    2011

  43. [43]

    Designing multi-label classifiers that maximize f measures: State of the art,

    I. Pillai, G. Fumera, and F. Roli, “Designing multi-label classifiers that maximize f measures: State of the art,” Pattern Recognition, vol. 61, pp. 394 – 404, 2017

    2017