Pith. sign in

REVIEW 4 major objections 6 minor 2 cited by

A Step-by-Step Guide to Creating a Robust Autonomous Drone Testing Pipeline

T0 review · 4 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A four-stage pipeline—software simulation, hardware-in-the-loop, controlled flights, and in-field trials—is presented as the route to safe autonomous drone deployment.

desk verdict A useful, clearly written tutorial on staged drone testing, with no new research results and a safety claim that outruns its self-reported case study. read the letter →

arxiv 2506.11400 v1 pith:RDYHQOAQ submitted 2025-06-13 cs.SE cs.RO

Yupeng Jiang , Yao Deng , Sebastian Schroder , Linfeng Liang , Suhaas Gambhir , Alice James , Avishkar Seth , James Pirrie
show 2 more authors
Yihao Zhang Xi Zheng
This is my paper
classification cs.SEcs.RO
keywords autonomousdronesdronetestingpipelinesoftware-in-the-loophardware-in-the-loopreal-worldmarker-basedlandingsimulation-basedtest-drivendevelopment
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Autonomous drones need a testing route that catches failures before they cause damage. This paper argues that a four-stage pipeline—software-in-the-loop (SIL) simulation, hardware-in-the-loop (HIL) testing, controlled real-world testing, and in-field testing—is the dependable way to validate such systems, and it gives step-by-step instructions for building each stage. The claim is carried by a running case study: a marker-based autonomous landing system that is iteratively tested and improved across all four stages. If the paper is right, drone teams that follow the pipeline should find integration bugs early, reduce deployment risk, and prepare missions for regulatory and real-world conditions. The paper frames this as practical guidance rather than a formal theorem.

What carries the argument

The carrying object is the testing pipeline itself: four stages that progressively substitute real hardware and real environment for simulation. SIL (software-in-the-loop) runs the drone's software against a high-fidelity simulator such as AirSim; HIL (hardware-in-the-loop) keeps the simulated world but replaces the flight controller and companion computer with physical hardware; controlled real-world testing moves to indoor arenas or netted enclosures; in-field testing flies in open operational environments. The marker-based autonomous landing system is the concrete mechanism tying the stages together: each stage exposes failures such as marker loss under glare, latency from a non-optimized inference stack, or wind-induced drift, and the paper shows how those failures drive iterative improvements. The named components that carry the demonstration are the ArUco marker detector, the ROS modular architecture, PX4/Pixhawk flight control, and a Jetson Nano with TensorRT inference.

What would settle it

Compare failure rates of teams that follow the full four-stage pipeline against teams that go straight to field testing. If drones that pass SIL and HIL fail in their first field missions at the same rate as untested drones, or if the measured SIL/HIL-to-field success correlation is near zero across several missions, the claim that the pipeline minimizes deployment risk is false.

Watch

Extended reading notes

Core claim

The paper's central claim is that a systematic four-stage pipeline, moving from SIL simulation to HIL validation to controlled indoor flight to open in-field trials, provides comprehensive validation and minimizes deployment risk for autonomous drones. Each stage catches a different class of failure: SIL verifies perception, planning, and control logic under repeatable simulation; HIL exposes hardware-software integration issues, latency, and thermal or inference bottlenecks; controlled real-world testing reveals physical and environmental effects in a safe setting; and in-field testing validates end-to-end behavior under wind, GPS variability, and changing light. The marker-based autonomous landing system is the evidence running through the paper: three generations of that system improved failure rates in SIL, an optimized HIL configuration achieved landing accuracy within ±5 cm, and field trials revealed wind-induced drift that was then reduced by retuning the controller. The conclusion restates the claim: following this pipeline prepares drones for safe and reliable real-world operations.

Load-bearing premise

The pipeline's promise of comprehensive validation rests on the assumption that defects and behaviors observed in SIL and HIL predict real-world performance well enough that passing those stages justifies field deployment.

Editorial extensions

If this is right

  • Defects that only appear when real hardware is in the loop, such as inference latency, thermal limits, and interface mismatches, can be caught before field trials.
  • Repeated SIL testing can drive design upgrades: the marker-based landing system's three generations show progressively lower collision and failure rates as detection and planning modules are replaced.
  • Controlled real-world tests expose environmental effects that simulation misses, such as wind-induced landing drift, enabling targeted retuning of stabilization parameters.
  • Teams progressing through the staged pipeline should be better positioned to document validation evidence and meet regulatory requirements before open-environment deployment.
  • Adopting neurosymbolic test generation, co-simulation, and digital twins is presented as the natural next step to cover edge cases that current SIL and HIL stages miss.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper: the pipeline's risk-reduction claim is testable by comparing field failure rates of teams that follow all four stages with teams that skip straight to flight; the paper does not supply that comparison.
  • Beyond the paper: the same staged logic could plausibly transfer to other modular cyber-physical systems, but the paper does not show that transferability outside drones.
  • Beyond the paper: the digital-twin trend implies that simulation evidence might eventually count toward certification, dissolving the boundary between testing and operation; the paper does not make that regulatory argument.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. This paper presents a step-by-step guide to building a four-stage testing pipeline for autonomous drones: software-in-the-loop (SIL) simulation, hardware-in-the-loop (HIL) testing, controlled real-world testing, and in-field testing. It includes setup instructions for AirSim and ROS, MAVROS/PX4 HIL integration, indoor and outdoor test procedures, and a discussion of future trends (neurosymbolic/LLM-based testing, co-simulation, digital twins). The guide is illustrated throughout with a marker-based autonomous landing system (MLS) used as a running example. The abstract and conclusion claim that following this pipeline enables developers to 'achieve comprehensive validation, minimize deployment risks, and prepare autonomous drones for safe and reliable real-world operations.'

Significance. The paper's practical value lies in its detailed, executable guidance: concrete commands, package layouts, launch files, and a demonstrated case study. If the central claim were supported, the paper would be a useful resource for drone teams, consolidating common practice into a single structured workflow. The authors deserve credit for the level of operational detail and for explicitly addressing safety and regulatory steps. However, the evidence adduced for the pipeline's effectiveness is limited: case-study results are self-reported and lack trial-level statistics, and the paper's own limitations sections acknowledge that simulation-to-field transfer is not well characterized. The work is better described as a practice-oriented tutorial than as a validated methodology.

major comments (4)
  1. [§2.1.2, §2.2.1, and Conclusion] The abstract and conclusion assert that the pipeline yields 'comprehensive validation' and 'minimized deployment risks.' This claim depends on the premise that passing SIL and HIL stages predicts real-world performance. The paper's own text states the opposite tendency: §2.1.2 says simulator limitations 'reduce predictive accuracy and restrict the transferability of insights from simulation to field deployment,' and §2.2.1 says simulated sensor noise profiles 'often do not precisely match those observed in actual drone hardware.' Neither the case study nor any other evidence quantifies these gaps. As written, the conclusion overstates what the paper demonstrates. Please either supply empirical evidence of transfer validity (e.g., comparing HIL pass/fail and field outcomes across multiple failures) or soften the headline claim to describe the pipeline as a recommended practice rather than a validated guarantee.
  2. [§2.2.2 and §2.3.2] The landing accuracy values '±5 cm' (indoor) and '±15 cm' (outdoor) are presented as results, but no methodology is given: number of trials, environmental conditions, ground-truth measurement (motion capture? manual?), and variance are all unspecified. Without these details, the numbers do not constitute reproducible evidence. The authors should report the trial data or present the figures only as qualitative observations.
  3. [§3.1] The NeuroStrata example is introduced as a 'representative example' of neurosymbolic testing, based on reference [76], a preprint with substantial author overlap. No experiments are described in this paper. If the future-trends section is intended as speculation, the text should say so explicitly rather than implying that these methods are currently performing the described functions.
  4. [§2.1.2 case study] The SIL case study reports MLS-V1 to V3 improvements (failure rates, collision rates) without any quantitative data: no successful trial counts, failure counts, or error metrics per version. For a guide whose central claim is risk reduction, these anchor results need to be available or explicitly marked as illustrative rather than measured.
minor comments (6)
  1. [§2.2.1] The heading 'Install MA VROS' contains a typo; it should read 'Install MAVROS.'
  2. [Appendix D] `marker_detector.py` uses the deprecated `cv2.aruco.Dictionary_get` and `cv2.aruco.DetectorParameters_create` APIs; in current OpenCV versions these have been replaced by `cv2.aruco.getPredefinedDictionary` and `cv2.aruco.DetectorParameters`. Also, `self.camera_matrix` is set as a Python list and later converted with `np.array`, but the list contains nested lists which should be explicitly cast to `np.float32` to avoid dtype issues in `estimatePoseSingleMarkers`.
  3. [§2.1.1] The ROS installation instructions use `apt-key add`, which is deprecated, and download the ROS signing key from a `master` branch URL that may change; consider recommending the current keyring procedure.
  4. [§2.3.2] The phrase 'consistently reaching within ±5 cm ... across multiple trials' would benefit from stating the number of trials and the observed range, since 'consistently' is otherwise undefined.
  5. [§2.4.3] 'maximum flight altitude (limited to 10–12 meters)' is ambiguous; if the range is 10 to 12 meters, please write '10 to 12 meters.'
  6. [Figures 5 and 6] Figures 5 and 6 appear to be photographs from the authors' own test campaign; if previously published in [61], that source should be credited in the captions.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the paper is a practical guide with no fitted predictions; self-citations are illustrative, not load-bearing.

full rationale

This paper is a practical engineering guide rather than a formal derivation, so there is no equation chain in which a fitted parameter or defined quantity is renamed as a prediction. The four-stage SIL-HIL-controlled-in-field sequence is presented as standard practice and is grounded in external references (AirSim, Gazebo, PX4, ArduPilot, Horri et al., the Harbin motion-capture testbed, Waymo/Baidu road-testing data) alongside the authors' marker-landing case study from refs [40,61]. The case study is self-reported, and the paper itself admits in Section 2.1.2 that simulator limitations 'reduce predictive accuracy and restrict the transferability of insights from simulation to field deployment' and in Section 2.2.1 that 'simulated sensor noise profiles often do not precisely match those observed in actual drone hardware'; these admissions weaken empirical support for the headline claim but are correctness and evidence concerns, not circularity. The NeuroStrata discussion [76] is a co-author self-citation, but it sits in the future-trends section and is not used to justify the core pipeline claim. No equation, fitted value, or definition reduces the conclusion to its inputs, so no circular step can be exhibited.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The paper is a tutorial, so there are no fitted parameters or invented entities. The central claims rest on domain assumptions about the adequacy of the staged pipeline, the transferability of simulation results, and the reliability of self-cited case-study data. These assumptions are partially acknowledged in the paper's own limitations sections.

assumptions (3)
  • domain assumption The four-stage pipeline (SIL, HIL, controlled real-world, in-field) is an appropriate and sufficient validation sequence for autonomous drones.
    The entire guide is built on this staging, which is presented in Section 1.3 as the de facto approach, but the paper does not demonstrate that this sequence is necessary or complete.
  • domain assumption Behavior observed in simulation and HIL transfers to real-world field performance.
    The claim that the pipeline minimizes deployment risk depends on this transfer. Section 2.1.2 notes simulators often fall short in multi-agent interactions and sensor fidelity, and Section 2.2.1 says simulated noise profiles do not precisely match real hardware, so transferability is acknowledged as limited.
  • domain assumption The self-cited case-study measurements from refs [40] and [61] are accurate and methodologically sound.
    The paper repeats quantitative outcomes such as ±5 cm and ±15 cm landing accuracy and the MLS-V1/V2/V3 failure progression without providing the underlying data or protocols.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Step-by-Step Guide to Creating a Robust Autonomous Drone Testing Pipeline." pith.science (2026). https://pith.science/paper/RDYHQOAQ

@misc{pith2026250611400,
  author       = {Pith},
  title        = {Pith review of: A Step-by-Step Guide to Creating a Robust Autonomous Drone Testing Pipeline},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RDYHQOAQ}},
  note         = {Machine review of arXiv:2506.11400}
}
read the original abstract

Autonomous drones are rapidly reshaping industries ranging from aerial delivery and infrastructure inspection to environmental monitoring and disaster response. Ensuring the safety, reliability, and efficiency of these systems is paramount as they transition from research prototypes to mission-critical platforms. This paper presents a step-by-step guide to establishing a robust autonomous drone testing pipeline, covering each critical stage: Software-in-the-Loop (SIL) Simulation Testing, Hardware-in-the-Loop (HIL) Testing, Controlled Real-World Testing, and In-Field Testing. Using practical examples, including the marker-based autonomous landing system, we demonstrate how to systematically verify drone system behaviors, identify integration issues, and optimize performance. Furthermore, we highlight emerging trends shaping the future of drone testing, including the integration of Neurosymbolic and LLMs, creating co-simulation environments, and Digital Twin-enabled simulation-based testing techniques. By following this pipeline, developers and researchers can achieve comprehensive validation, minimize deployment risks, and prepare autonomous drones for safe and reliable real-world operations.

Figures

Figures reproduced from arXiv: 2506.11400 by the authors.

Figure 1
Figure 1. Components of a Popular Modular Autonomous Drone System [61]. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Overview of the autonomous drone testing pipeline. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Overview of the marker-based autonomous landing process. [PITH_FULL_IMAGE:figures/full_fig_p014_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Illustrative failures observed during SIL simulation testing of a marker-based autonomous landing [PITH_FULL_IMAGE:figures/full_fig_p016_4.png]
Figure 5
Figure 5. Figure 5: Fully assembled autonomous drone platform. [PITH_FULL_IMAGE:figures/full_fig_p025_5.png]
Figure 6
Figure 6. Figure 6: Controlled indoor testing setup for validating the marker-based landing system. [PITH_FULL_IMAGE:figures/full_fig_p025_6.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Validating Terrain Models in Digital Twins for Trustworthy sUAS Operations

    cs.SE 2025-08 conditional novelty 4.0 of 10

    A three-axis validation framework (unit to acceptance, simulation to real flight, simple to harsh terrain) is demonstrated on a drone terrain model, with geolocation errors of 1.5-4.2m and over 10m under poor GPS.

  2. Regression Testing Optimization for ROS-based Autonomous Systems: A Comprehensive Review of Techniques

    cs.SE 2025-06 conditional novelty 4.0 of 10

    A survey that categorizes 122 papers on regression test optimization and argues ROS-based autonomous systems need new semantic and neurosymbolic approaches.

Reference graph

Works this paper leans on

80 extracted references · 71 canonical work pages · cited by 2 Pith papers

  1. [76]

    Xi Zheng, Ziyang Li, Ivan Ruchkin, Ruzica Piskac, and Miroslav Pajic. 2025. NeuroStrata: Harnessing Neurosymbolic Paradigms for Improved Design, Testability, and Verifiability of Autonomous CPS.ArXivabs/2502.12267 (2025)

    arXiv 2025

  2. [1]

    Afsoon Afzal, Deborah S Katz, Claire Le Goues, and Christopher S Timperley. 2020. A study on the challenges of using robotics simulators for testing.arXiv preprint arXiv:2004.07368(2020)

    arXiv 2020

  3. [2]

    Nourdine Aliane. 2024. A Survey of Open-Source UAV Autopilots.Electronics13, 23 (2024), 4785

    work page 2024

  4. [3]

    Yagna SH Annadata, Aiswariya Thazhathethil, Vishnuvaradhan Moganarengam, and Tooraj Nikoubin. 2025. Enhancing Drone-Based Precision Agriculture: Performance Optimization of TinyML Models on Edge Devices and Adaptive Path Planning.SN Computer Science6, 2 (2025), 132

    work page 2025

  5. [4]

    Angelos Antonopoulos, Michail G Lagoudakis, and Panagiotis Partsinevelos. 2022. A ROS multi-tier UAV localization module based on GNSS, inertial and visual-depth data.Drones6, 6 (2022), 135

    work page 2022

  6. [5]

    ApolloAuto. 2023. Apollo. https://github.com/ApolloAuto/apollo

    work page 2023

  7. [6]

    ArduPilot. 2025. ArduPilot. https://github.com/ArduPilot/ardupilot

    work page 2025

  8. [7]

    ArduPilot. 2025. ArduPilot - Versatile, Trusted, Open. https://ardupilot.org/

    work page 2025

Show all 80 references
  1. [8]

    PX4 Autopilot. 2025. PX4 Autopilot User Guide. https://docs.px4.io/main/en/

    2025

  2. [9]

    AutowareFoundation. 2025. Autoware. https://github.com/autowarefoundation/autoware

    2025

  3. [10]

    Thom Badings, Licio Romao, Alessandro Abate, and Nils Jansen. 2023. Probabilities are not enough: Formal controller synthesis for stochastic dynamical models with epistemic uncertainty. InProceedings of the AAAI Conference on Artificial Intelligence, Vol. 37. 14701–14710

    2023

  4. [11]

    Benewake. 2025. TFmini Plus 12 m IP65 Protected LiDAR. https://en.benewake.com/TFminiPlus/index.html

    2025

  5. [12]

    Gary Bradski. 2000. The opencv library.Dr. Dobb’s Journal: Software Tools for the Professional Programmer25, 11 (2000), 120–123

    2000

  6. [13]

    Flavia Causa and Giancarmine Fasano. 2021. Multiple UAVs trajectory generation and waypoint assignment in urban environment based on DOP maps.Aerospace Science and Technology110 (2021), 106507. A Step-by-Step Guide to Creating a Robust Autonomous Drone Testing Pipeline 33

    2021

  7. [14]

    Chenyi Chen, Ari Seff, Alain Kornhauser, and Jianxiong Xiao. 2015. Deepdriving: Learning affordance for direct perception in autonomous driving. InProceedings of the IEEE international conference on computer vision. 2722–2730

    2015

  8. [15]

    DHL Freight Connections. 2024. Future of Freight: Use of Drones in Logistics. https://dhl-freight-connections.com/en/ solutions/future-of-freight-the-use-of-drones-in-logistics/

    2024

  9. [16]

    CUAV. 2025. CUAV X7+ Pro Pixhawk For PX4 APM Controller. https://store.cuav.net/shop/x7-pro/

    2025

  10. [17]

    CUAV. 2025. NEO 3 Global Satellite Navigation Module. https://www.cuav.net/en/neo-3-en/

    2025

  11. [18]

    Yao Deng, Xi Zheng, Mengshi Zhang, Guannan Lou, and Tianyi Zhang. 2022. Scenario-based test reduction and prioritization for multi-module autonomous driving systems. InProceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations o...

    2022

  12. [19]

    Yao Deng, Xi Zheng, Tianyi Zhang, Huai Liu, Guannan Lou, Miryung Kim, and Tsong Yueh Chen. 2023. A Declarative Metamorphic Testing Framework for Autonomous Driving.IEEE Transactions on Software Engineering49 (2023), 1964–1982

    2023

  13. [20]

    California DMV. 2024. Disengagement Reports. https://www.dmv.ca.gov/portal/vehicle-industry-services/autonomous- vehicles/disengagement-reports/

    2024

  14. [21]

    Alexey Dosovitskiy, German Ros, Felipe Codevilla, Antonio Lopez, and Vladlen Koltun. 2017. CARLA: An open urban driving simulator. InConference on robot learning. PMLR, 1–16

    2017

  15. [22]

    EASA. 2025. Drone Regulatory System. https://www.easa.europa.eu/en/domains/drones-air-mobility/drones-air- mobility-landscape/Understanding-European-Drone-Regulations-and-the-Aviation-Regulatory-System

    2025

  16. [23]

    Elham Fakhraian, Ivana Semanjski, Silvio Semanjski, and El-Houssaine Aghezzaf. 2023. Towards safe and efficient unmanned aircraft system operations: Literature review of digital twins’ applications and european union regulatory compliance.Drones7, 7 (2023), 478

    2023

  17. [24]

    Epic Games. 2025. Unreal Engine. https://www.unrealengine.com

    2025

  18. [25]

    Sergio Garrido-Jurado, Rafael Muñoz-Salinas, Francisco José Madrid-Cuevas, and Manuel Jesús Marín-Jiménez. 2014. Automatic generation and detection of highly reliable fiducial markers under occlusion.Pattern Recognition47, 6 (2014), 2280–2292

    2014

  19. [26]

    Gazebo. 2025. Gazebo. https://gazebosim.org

    2025

  20. [27]

    Cláudio Gomes, Casper Thule, David Broman, Peter Gorm Larsen, and Hans Vangheluwe. 2018. Co-simulation: a survey.ACM Computing Surveys (CSUR)51, 3 (2018), 1–33

    2018

  21. [28]

    GreyB. 2024. Precision Agriculture Drones: An AgriTech Trend in 2024. https://www.greyb.com/blog/precision- agriculture-drones/

    2024

  22. [29]

    Johvany Gustave, Jamy Chahal, and Assia Belbachir. 2020. Functional Architecture using ROS for Autonomous UAVs.. InICINCO. 506–512

    2020

  23. [30]

    Peter E Hart, Nils J Nilsson, and Bertram Raphael. 1968. A formal basis for the heuristic determination of minimum cost paths.IEEE transactions on Systems Science and Cybernetics4, 2 (1968), 100–107

    1968

  24. [31]

    HIGHTOPO. 2025. Low-Altitude Economy Management Platform - Drone Delivery Management Sys- tem. https://hightopo.medium.com/low-altitude-economy-management-platform-drone-delivery-management- system-0edc30247506

    2025

  25. [32]

    Armin Hornung, Kai M Wurm, Maren Bennewitz, Cyrill Stachniss, and Wolfram Burgard. 2013. OctoMap: An efficient probabilistic 3D mapping framework based on octrees.Autonomous robots34 (2013), 189–206

    2013

  26. [33]

    Nadjim Horri and Mikolaj Pietraszko. 2022. A tutorial and review on flight control co-simulation using matlab/simulink and flight simulators.Automation3, 3 (2022), 486–510

    2022

  27. [34]

    Intel. 2025. Depth Camera D435. https://www.intelrealsense.com/depth-camera-d435/

    2025

  28. [35]

    Maxwell J Jacobson and Yexiang Xue. 2025. Integrating symbolic reasoning into neural generative models for design generation.Artificial Intelligence339 (2025), 104257

    2025

  29. [36]

    Hyeong Ryeol Kam, Sung-Ho Lee, Taejung Park, and Chang-Hun Kim. 2015. Rviz: a toolkit for real domain data visualization.Telecommunication Systems60 (2015), 337–345

    2015

  30. [37]

    Sertac Karaman and Emilio Frazzoli. 2011. Sampling-based algorithms for optimal motion planning.The international journal of robotics research30, 7 (2011), 846–894

    2011

  31. [38]

    Hemani Kaushal and Georges Kaddoum. 2016. Optical communication in space: Challenges and mitigation techniques. IEEE communications surveys & tutorials19, 1 (2016), 57–96

    2016

  32. [39]

    Sajad Khatiri, Sebastiano Panichella, and Paolo Tonella. 2023. Simulation-based test case generation for unmanned aerial vehicles in the neighborhood of real flights. In2023 IEEE Conference on Software Testing, Verification and Validation (ICST). IEEE, 281–292

    2023

  33. [40]

    Linfeng Liang, Yao Deng, Kye Morton, Valtteri Kallinen, Alice James, Avishkar Seth, Endrowednes Kuantama, Subhas Mukhopadhyay, Richard Han, and Xi Zheng. 2025. GARL: Genetic Algorithm-Augmented Reinforcement Learning to Detect Violations in Marker-Based Autonomous Landing Syst...

    2025

  34. [41]

    Tom Looman. 2024. Unreal Engine 5.5 Performance Highlights. https://www.tomlooman.com/unreal-engine-5-5- performance-highlights/

    2024

  35. [42]

    Guannan Lou, Yao Deng, Xi Zheng, Mengshi Zhang, and Tianyi Zhang. 2022. Testing of autonomous driving systems: where are we and where should we go?. InProceedings of the 30th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engine...

    2022

  36. [43]

    MathWorks. 2025. Simulink. https://www.mathworks.com/products/simulink.html

    2025

  37. [44]

    MAVROS. 2025. MAVROS. https://github.com/MAVLink/MAVROS

    2025

  38. [45]

    Lorenz Meier, Dominik Honegger, and Marc Pollefeys. 2015. PX4: A node-based multithreaded open source robotics framework for deeply embedded platforms. In2015 IEEE international conference on robotics and automation (ICRA). IEEE, 6235–6240

    2015

  39. [46]

    Mayank Mittal, Rohit Mohan, Wolfram Burgard, and Abhinav Valada. 2019. Vision-based autonomous UAV navigation and landing for urban search and rescue. InThe International Symposium of Robotics Research. Springer, 575–592

    2019

  40. [47]

    UAV Navigation. 2024. VECTOR-HIL: An Essential Tool for the Development and Validation of UAVs in Controlled Environments. https://www.uavnavigation.com/company/blog/vector-hil-essential-tool-development-and-validation- uavs-controlled-environments

    2024

  41. [48]

    Gorilla Netting. 2025. Netted Drone Enclosures. https://www.gorillanetting.com/services/commercial-netting/netted- drone-enclosures/

    2025

  42. [49]

    PR Newswire. 2021. Baidu Tops Beijing Autonomous Vehicles Road Test Report for the Third Consecutive Year. https://www.prnewswire.com/news-releases/baidu-tops-beijing-autonomous-vehicles-road-test-report-for-the- third-consecutive-year-301222929.html

    2021

  43. [50]

    NOKOV. 2025. Indoor Motion Capture-Based UAV Flight Control Testing Platform. https://www.nokov.com/support/ case_studies_detail/Laboratory-Experiment-System-of-Multi-Rotor-UAV-Based-on-Motion-Capture-System.html

    2025

  44. [51]

    NVIDIA. 2025. NVIDIA Jetson Nano. https://www.nvidia.com/en-au/autonomous-machines/embedded-systems/jetson- nano/product-development/

    2025

  45. [52]

    NVIDIA. 2025. TensorRT. https://github.com/NVIDIA/TensorRT

    2025

  46. [53]

    NVIDIA. 2025. What Is a Digital Twin? https://www.nvidia.com/en-us/glossary/digital-twin/

    2025

  47. [54]

    Pixhawk. [n. d.]. Holybro Pixhawk 4 Mini. https://docs.px4.io/main/en/flight_controller/pixhawk4_mini.html

  48. [55]

    Morgan Quigley, Ken Conley, Brian Gerkey, Josh Faust, Tully Foote, Jeremy Leibs, Rob Wheeler, Andrew Y Ng, et al

  49. [56]

    Adil Rasheed, Omer San, and Trond Kvamsdal. 2020. Digital twin: Values, challenges and enablers from a modeling perspective.IEEE access8 (2020), 21980–22012

    2020

  50. [57]

    Guodong Rong, Byung Hyun Shin, Hadi Tabatabaee, Qiang Lu, Steve Lemke, M¯artin, š Možeiko, Eric Boise, Geehoon Uhm, Mark Gerow, Shalin Mehta, et al. 2020. Lgsvl simulator: A high fidelity simulator for autonomous driving. In 2020 IEEE 23rd International conference on intellige...

    2020

  51. [58]

    ROS. 2025. Understanding real-time programming. https://docs.ros.org/en/foxy/Tutorials/Demos/Real-Time- Programming.html

    2025

  52. [59]

    Emrullah ŞAHiN, Naciye Nur Arslan, and Durmuş Özdemir. 2025. Unlocking the black box: an in-depth review on interpretability, explainability, and reliability in deep learning.Neural Computing and Applications37, 2 (2025), 859–965

    2025

  53. [60]

    Mrinmoy Sarkar, Xuyang Yan, Shamila Nateghi, Bruce J Holmes, Kyriakos G Vamvoudakis, and Abdollah Homaifar

  54. [61]

    Sebastian Schroder, Yao Deng, Alice James, Avishkar Seth, Kye Morton, Subhas Mukhopadhyay, Richard Han, and Xi Zheng. 2025. Towards Robust Autonomous Landing Systems: Iterative Solutions and Key Lessons Learned.arXiv preprint arXiv:2505.12176(2025)

    2025 arXiv

  55. [62]

    2015.Smart autonomous aircraft: flight control and planning for UA V

    Yasmina Bestaoui Sebbane. 2015.Smart autonomous aircraft: flight control and planning for UA V. Crc Press

    2015

  56. [63]

    Shital Shah, Debadeepta Dey, Chris Lovett, and Ashish Kapoor. 2018. Airsim: High-fidelity visual and physical simulation for autonomous vehicles. InField and Service Robotics: Results of the 11th International Conference. Springer, 621–635

    2018

  57. [64]

    Hazim Shakhatreh, Ahmad H Sawalmeh, Ala Al-Fuqaha, Zuochao Dou, Eyad Almaita, Issa Khalil, Noor Shamsiah Othman, Abdallah Khreishah, and Mohsen Guizani. 2019. Unmanned aerial vehicles (UAVs): A survey on civil applications and key research challenges.Ieee Access7 (2019), 48572–48634

    2019

  58. [65]

    Nabil Shaukat, Shival Dubey, Bilal Kaddouh, Andy Blight, Lenka Mudrich, Pedro Ribeiro, Hugo Araujo, Rob Richardson, Louise Dennis, Ana Cavalcanti, et al. 2024. Trustworthy ROS Software Architecture for Autonomous Drones Missions: From RoboChart Modelling to ROS Implementation....

    2024

  59. [66]

    Ioan A Sucan, Mark Moll, and Lydia E Kavraki. 2012. The open motion planning library.IEEE Robotics & Automation Magazine19, 4 (2012), 72–82

    2012

  60. [67]

    Simon Suo, Sebastian Regalado, Sergio Casas, and Raquel Urtasun. 2021. Trafficsim: Learning to simulate realistic multi-agent behaviors. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 10400– 10409

    2021

  61. [68]

    Fei Tao, Qinglin Qi, Lihui Wang, and AYC Nee. 2019. Digital twins and cyber–physical systems toward smart manufacturing and industry 4.0: Correlation and comparison.Engineering5, 4 (2019), 653–661

    2019

  62. [69]

    TWAICE. 2025. This software determines how healthy expensive batteries still are. https://www.twaice.com/newsroom/ twaice-software

    2025

  63. [70]

    András Varga and Rudolf Hornig. 2008. An overview of the OMNeT++ simulation environment. InProceedings of the 1st international conference on Simulation tools and techniques for communications, networks and systems & workshops. 1–10

    2008

  64. [71]

    Waymo. 2025. Waymo - Self-Driving Cars - Autonomous Vehicles - Ride-Hail. https://waymo.com/

    2025

  65. [72]

    Wired. 2022. Drones Have Transformed Blood Delivery in Rwanda. https://www.wired.com/story/drones-have- transformed-blood-delivery-in-rwanda/

    2022

  66. [73]

    HAWK’S WORK. 2025. F450 Drone Kit. https://www.hawks-work.com/products/f450-drone-kit-to-build-diy-450mm- wheelbase-4-axis-multi-rotor-drone-kit-d

    2025

  67. [74]

    Smart Drones World. 2024. 10 Stunning Examples of Drone Aerial Photography. https://smart-drones-world.com/ reviews/10-stunning-examples-of-drone-aerial-photography/

    2024

  68. [75]

    Bohan Zhang and Ankit Agrawal. 2024. DroneWiS: Automated Simulation Testing of small Unmanned Aerial System in Realistic Windy Conditions. InProceedings of the 39th IEEE/ACM International Conference on Automated Software Engineering. 2358–2361

    2024

  69. [77]

    Xin Zhou, Zhepei Wang, Hongkai Ye, Chao Xu, and Fei Gao. 2020. Ego-planner: An esdf-free gradient-based local planner for quadrotors.IEEE Robotics and Automation Letters6, 2 (2020), 478–485

    2020

  70. [78]

    drone_name

    Xingkui Zhu, Shuchang Lyu, Xu Wang, and Qi Zhao. 2021. TPH-YOLOv5: Improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios. InProceedings of the IEEE/CVF international conference on computer vision. 2778–2788. 36 Jiang et al. A Pac...

    2021

  71. [2009]

    InICRA workshop on open source software, Vol

    ROS: an open-source Robot Operating System. InICRA workshop on open source software, Vol. 3. Kobe, 5

  72. [2022]

    InIntelligent Systems and Applications: Proceedings of the 2021 Intelligent Systems Conference (IntelliSys) Volume 1

    A framework for testing and evaluation of operational performance of multi-UAV systems. InIntelligent Systems and Applications: Proceedings of the 2021 Intelligent Systems Conference (IntelliSys) Volume 1. Springer, 355–374

    2021

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.