Pith. sign in

REVIEW 3 major objections 6 minor 37 references

Multi-robot Aerial Soft Manipulator For Floating Litter Collection

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A pair of drones connected by a hook-tipped rope can autonomously collect floating litter in still, narrow waterways.

desk verdict Real outdoor two-UAV rope manipulator for litter pickup with a genuinely adaptive planner, but the quantitative ablation is underpowered and the water-channel text has a w=1.0/w=1.5 inconsistency. read the letter →

arxiv 2507.03517 v1 pith:MGPUYLI6 submitted 2025-07-04 cs.RO cs.SYeess.SY

Antonio González-Morgado , Sander Smits , Guillermo Heredia , Anibal Ollero , Alexandre Krupa , François Chaumette , Fabien Spindler , Antonio Franchi
show 1 more author
Chiara Gabellieri
This is my paper
classification cs.ROcs.SYeess.SY
keywords aerialmanipulationmulti-robotsystemsfloatinglittercollectionsoftmanipulatorvisualservoingropeshapeplanningparabolamodeloutdoorexperiments
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

The paper argues that a flexible rope suspended between two aerial robots can serve as a soft manipulator for collecting floating litter, and that the rope's shape, not just the robot trajectories, is the controllable quantity that decides whether a grasp succeeds. It presents the complete control loop: an offline planner chooses a parabolic rope shape that widens near the litter to maximize hook coverage and narrows elsewhere to reduce rope tension; an onboard depth camera estimates the rope as a parabola; and a shape visual servoing controller corrects both robot trajectories symmetrically so the hook at the rope's midpoint stays on its planned path. Outdoor experiments in a grass field and a real water channel report successful grasps, and the ablation study shows the adaptive widening mechanism raises success from 0–16.7 percent at the lowest planner weight to 66.6–83.3 percent at the highest. If this holds, two small drones and a cheap rope become a practical, low-cost complement to fixed barriers and boat collectors in still, narrow waterways.

What carries the argument

The load-bearing mechanism is the adaptive rope-shape planner coupled to a parabola-based visual servoing loop. The planner solves for the rope curvature $a$ and the drone separation $d$ under the constraint that the parabola arc length equals the rope length $l_{rope}$, minimizing an objective $J = f_t + w_{gr} f_{gr}$ in which $f_t$ is rope tension and $f_{gr}$ is unused hook width; the weight $w_{gr}$ is a sigmoid of the distance from the hook to the litter, so the grasping term is active only near the target. The shape is represented as $z = a y^2 + b y$ in the rope frame, and the visual servoing law converts the error in $(a_p, b_p, \psi_p)$ into a relative velocity through an interaction matrix, with the correction split symmetrically between the two drones to keep the rope midpoint on its planned trajectory. This combination is what lets the system widen the rope near the litter, keep tension low elsewhere, and still hold the hook on the desired collection path.

What would settle it

Run the water-channel collection with a crosswind or with the hook snagged on a submerged branch while the visual servoing is active, and measure the rope-plane roll angle from the onboard point cloud: if the estimated roll moves substantially away from zero and the rope midpoint deviates from the planned hook path, the fixed-vertical-plane model is the point of failure.

Watch

Extended reading notes

Core claim

The central claim is that a two-UAV system connected by a rope with a hook-based end-effector can autonomously collect floating litter in still, narrow water bodies, provided the rope shape is planned as an optimization and then regulated by visual feedback. The planner minimizes rope tension while also minimizing the unused hook width, with the second term switched on by a sigmoid only when the rope's midpoint is near the litter; this makes the robots spread apart as they approach the target and stay close together elsewhere. The rope is modeled as a parabola, and a shape visual servoing controller converts the error in the parabola parameters into a relative velocity correction that is split equally and oppositely between the two robots, so the midpoint trajectory of the hook is preserved. In the grass-field ablation, higher grasping weight raised success rates from 0 and 16.7 percent (low weight) to 66.6 and 83.3 percent (high weight) on straight and circular approaches; in the water channel, the adaptive planner collected at least one bag in four of six runs while the always-narrow rope succeeded in one of six.

Load-bearing premise

The whole control chain assumes the rope hangs in a vertical plane, with the roll angle of the rope plane near zero, and that the litter sits at a known, fixed spot; wind, currents, asymmetric drag, or a snagged object tilting that plane would invalidate the planned hook trajectory and the visual-servoing corrections.

Editorial extensions

If this is right

  • Autonomous litter collection in still channels and lakes no longer requires boats or fixed barriers: two small quadrotors and a rope can perform the grasp, and only takeoff and landing need human pilots.
  • The planner's adaptive weight gives operators a single tunable knob that trades grasping success against robot separation, flight height, and roll-angle limits, which is also a safety and collision-avoidance lever.
  • Because the visual servoing correction is split symmetrically, the hook path is preserved during shape regulation, so the same controller can be reused for other cable-shape tasks where a midpoint or end-effector trajectory must be kept.
  • The system can be made fully autonomous by replacing the assumed known litter position with online litter detection and tracking, and by replanning the hook trajectory when the litter moves.
  • Alternative end-effectors, such as nets, could extend the same adaptive rope-shape planner to rigid litter like bottles and cans.

Reading between the lines

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

  • Beyond the paper, the same two-robot symmetric correction could apply to other suspended-cable manipulation tasks, such as transporting a payload or guiding a hose, whenever the cable midpoint must stay on a reference path.
  • Beyond the paper, the sharp success-rate gap suggests the planner's widening does double duty: it increases hook coverage and it keeps propeller downwash away from the target, so the benefit should transfer to any aerial manipulation near loose or floating objects.
  • Beyond the paper, a natural stress test is moving water: because the assumption of a vertical rope plane is explicit, adding roll-angle feedback or a catenary model may be needed before the system can work in currents.
  • Beyond the paper, replacing the hook with a net and using the same planner would be a direct test of whether the adaptive shape, not the specific end-effector, is what drives grasping success.
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

3 major / 6 minor

Summary. The paper describes a multi-robot aerial soft manipulator for collecting floating litter from water surfaces. The system consists of two quadrotors connected by a flexible rope with a hook-based tool at the midpoint. The rope shape is modeled as a parabola, and an offline planner computes the desired parabola parameters by minimizing a weighted sum of rope tension and unused hook width, with the weight adapted via a sigmoid of the distance to the litter. A visual servoing controller using an RGB-D camera estimates the parabola parameters and generates symmetric position corrections for the two robots to track the planned shape while keeping the midpoint on its reference trajectory. The system is validated through outdoor experiments in a grass field and in a water channel, and an ablation study varying the planner weight w is presented. The paper claims successful grasping and that the adaptive planner mechanism improves the success rate.

Significance. The main contribution is an integrated system demonstration: two UAVs, a flexible rope, perception, planning, and control are combined and tested outdoors in a realistic water channel. The paper deserves credit for performing real experiments with wind and water reflections, and the symmetric control action that preserves the hook trajectory is a sensible design improvement over earlier leader-follower schemes. The shape-tracking plots appear consistent with the described behavior. However, the quantitative claim that the adaptive mechanism improves success rests on very small samples (six trials per condition) without statistical analysis, and a key modeling assumption (near-zero roll of the rope plane) is neither reported nor controlled. These issues prevent the current evidence from being conclusive, though the system concept is promising.

major comments (3)
  1. [Section II-E-2, Fig. 6] The ablation study compares success rates over six flights per condition. With n=6, the observed differences (e.g., 0/6 vs. 4/6 for straight trajectory with w=0 vs. w=1.5) are not statistically significant; no confidence intervals or tests are reported. The statement that 'higher weight values lead to improved grasping capabilities' is therefore stronger than the data support. Please provide exact binomial confidence intervals or a statistical test (e.g., Fisher's exact test), or temper the claim to a qualitative trend.
  2. [Section IV-B and IV-E] The planner and visual servoing controller rely on the assumption that the rope plane roll angle phi remains close to zero. Although phi_p is estimated via Eq. (15), the paper never reports its values during any experiment and no control action regulates it. Given that the hook tool is dragged through water and the trials were conducted in 2-3 m/s wind, this unverified assumption is load-bearing for the outdoor-validity claim. Report the measured phi_p time series or otherwise demonstrate that the roll remains small; otherwise the projected parabola used for feedback can be systematically biased.
  3. [Section II-E-3] The water-channel ablation contains an internal inconsistency: the text first states that w=1.0 is used to limit separation, then says that results for w=1.0 are compared with w=0, but reports a success rate 'for w=1.5 was 66.6%,' which exactly matches the grass-field straight-trajectory w=1.5 result. Please correct the weight label and the reported success rates; as written, the water-channel comparison is not interpretable.
minor comments (6)
  1. [Section II-A] The word 'refereed' should be 'referred'.
  2. [Fig. 8 caption] The caption lists time instants t=20, 25, 30 s, but the water-channel snapshots in Fig. 7 are at t=40-60 s; the caption needs correction.
  3. [Eq. (12)] Equation (12) has a formatting error: 'Z vvs_rel' should be an integral sign (∫ vvs_rel dt).
  4. [General] There are typos such as 'M atlab' and 'intrisic' that should be corrected.
  5. [Section II-E-3] The success metric in the water-channel experiment is 'at least one bag collected'; please define this criterion precisely and consider reporting the number of bags collected per trial.
  6. [Section II-E-2] The ablation is partly an internal consistency check because the planner's objective includes hook-covered width and the success metric is hook-based grasping; the paper would benefit from explicitly acknowledging this and emphasizing the independent evidence from shape-tracking errors.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the claimed outdoor validation and ablation results are empirical, and the self-citations are background/basis rather than load-bearing.

full rationale

The paper's central claims are empirical and do not reduce to their inputs by the paper's own equations. The adaptive planner maximizes hook-covered width (Eqs. 5-8) and the success metric is hook-based grasping, so the ablation is partly an internal consistency check, but no equation maps d_hook to observed success; the reported success rates are outcomes of repeated flights, and the authors also attribute the w=0 failures to downwash displacing the litter, a physical mechanism independent of the planner's cost. The shape visual servoing and parabola estimation are adapted from prior work [30],[31] by co-authors, and [23],[24] are motivational self-citations; these are not used to prove that the current system works, and the current system is tested outdoors, so the citations are not load-bearing. The unverified assumption that the rope-plane roll angle remains near zero (Sec. IV-B) is an experimental and robustness concern, not a circular derivation. No fitted parameter is relabeled as a prediction. Score 2 reflects only the presence of minor, non-load-bearing self-citations and the conceptual alignment between the planner objective and the success metric.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central engineering claims rest primarily on domain assumptions (parabola model, still-water known litter, negligible roll) and on a set of manually chosen tuning parameters (planner weights, safety limits, estimation and controller gains). No new physical entities are introduced.

free parameters (6)
  • w (grasping objective weight) = 0.0, 0.5, 1.0, 1.5; water channel uses 1.0
    User-selected weight in Eq. (8) balancing rope tension against hook coverage; the ablation study varies it and the water-channel setup selects 1.0 to limit robot separation.
  • kgr (sigmoid slope) = 1.0
    Chosen in Section II-E1; its selection is deferred to Supplementary Method 6.
  • kpos (sigmoid center) = 1.0
    Chosen in Section II-E1; determines the distance at which the grasping weight becomes active.
  • hmin and phi_max (safety and roll limits) = Not reported in main text
    Operator-defined limits in Eq. (4) that set the maximum drone separation dmax; no numerical values are given.
  • Estimation weights wm, wd, wp = Not reported in main text
    Weights in the parabola-fitting optimization Eq. (17), chosen based on sensor precision; exact values deferred to Supplementary Method 7.
  • PI gains kc, ki and window size nw = Not reported in main text
    Controller gains in Eq. (11) for the visual servoing law; no values are provided.
assumptions (5)
  • domain assumption The suspended rope is modeled as a parabola z = a y^2 + b y rather than a catenary.
    This model is used throughout planning (Eq. 2), control (Eqs. 10-11), and estimation (Eq. 17); the paper motivates it by robustness and low computational cost, but it is an approximation of the true static shape.
  • domain assumption Litter positions are known and fixed; the water is still.
    Section II-A states: 'We assume that the litter positions are known for the offline planner, which is reasonable under the hypothesis of still waters.' The entire offline plan depends on this.
  • domain assumption The roll angle of the rope plane remains close to zero.
    Section IV-B says 'we assume this angle remains close to zero'; roll is not controlled by the visual servoing law, so the vertical-plane parabola model requires this to hold throughout the task.
  • domain assumption The rope is in quasi-static equilibrium and the mass of collected litter is negligible compared to the rope.
    The tension objective Eq. (6) uses mrope and neglects litter mass; dynamic effects, added mass from captured litter, and rope elasticity are ignored in planning.
  • ad hoc to paper The adaptive sigmoid weighting function Eq. (8) is a suitable heuristic.
    The functional form of wgr and its parameters are chosen by the authors rather than derived; the paper tunes them and tests a few values.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multi-robot Aerial Soft Manipulator For Floating Litter Collection." pith.science (2026). https://pith.science/paper/MGPUYLI6

@misc{pith2026250703517,
  author       = {Pith},
  title        = {Pith review of: Multi-robot Aerial Soft Manipulator For Floating Litter Collection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MGPUYLI6}},
  note         = {Machine review of arXiv:2507.03517}
}
read the original abstract

Removing floating litter from water bodies is crucial to preserving aquatic ecosystems and preventing environmental pollution. In this work, we present a multi-robot aerial soft manipulator for floating litter collection, leveraging the capabilities of aerial robots. The proposed system consists of two aerial robots connected by a flexible rope manipulator, which collects floating litter using a hook-based tool. Compared to single-aerial-robot solutions, the use of two aerial robots increases payload capacity and flight endurance while reducing the downwash effect at the manipulation point, located at the midpoint of the rope. Additionally, we employ an optimization-based rope-shape planner to compute the desired rope shape. The planner incorporates an adaptive behavior that maximizes grasping capabilities near the litter while minimizing rope tension when farther away. The computed rope shape trajectory is controlled by a shape visual servoing controller, which approximates the rope as a parabola. The complete system is validated in outdoor experiments, demonstrating successful grasping operations. An ablation study highlights how the planner's adaptive mechanism improves the success rate of the operation. Furthermore, real-world tests in a water channel confirm the effectiveness of our system in floating litter collection. These results demonstrate the potential of aerial robots for autonomous litter removal in aquatic environments.

Figures

Figures reproduced from arXiv: 2507.03517 by the authors.

Figure 1
Figure 1. To achieve this, we propose the use of an aerial system [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 1
Figure 1. Floating litter collection with an aerial soft manipulator system. Two aerial robots, robot 1 and robot 2, control the shape of a flexible [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Rope shape for litter collection. a) Different rope shapes during the litter collection, including the needed force (red arrows) and the [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figures from the paper (8 more)
Figure 3
Figure 3. Figure 3: Control framework for litter collection, composed by the offline rope-shape planner (green box), the rope shape controller (pink box) f [PITH_FULL_IMAGE:figures/full_fig_p005_3.png]
Figure 4
Figure 4. Figure 4: Results of the experiments in the grass field. a) 3D and 2D views of the trajectories. The transparency effect indicates the temporal [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Parabola shape estimation during the validation experiment in the grass field. a) Onboard camera view (first column), filtered image [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Ablation study in the grass field for different trajectories and weights [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: Results of the experiments in water channel. a) 3D and 2D views of the trajectories during the floating litter collection in channel. [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: Parabola shape estimation during the the experiments in water channel. a) Onboard camera view (first column), filtered image (second [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 9
Figure 9. Figure 9: Aerial system for floating litter collection. a) Main components of the aerial system, including two commercial Holybro X500 [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]
Figure 10
Figure 10. Figure 10: Frames used in this work. a) Main frames and position vectors. b) Parabola equations for the [PITH_FULL_IMAGE:figures/full_fig_p012_10.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

37 extracted references · 36 canonical work pages

  1. [1]

    Solutions to plastic pollution,

    C. Horejs, “Solutions to plastic pollution,” Nature Reviews Materials , vol. 5, no. 9, pp. 641–641, 2020

    work page 2020

  2. [2]

    Global environmental plastics dispersal under oecd policy scenarios towards 2060,

    J. Sonke, A. Koenig, T. Segur, and N. Yakovenko, “Global environmental plastics dispersal under oecd policy scenarios towards 2060,” 2024

    work page 2024

  3. [3]

    Ridding our rivers of plastic: A framework for plastic pollution capture device selection,

    O. K. Helinski, C. J. Poor, and J. M. Wolfand, “Ridding our rivers of plastic: A framework for plastic pollution capture device selection,” Marine pollution bulletin , vol. 165, p. 112095, 2021

    work page 2021

  4. [4]

    Risk assessment of microplas- tic particles,

    A. A. Koelmans, P. E. Redondo-Hasselerharm, N. H. M. Nor, V . N. de Ruijter, S. M. Mintenig, and M. Kooi, “Risk assessment of microplas- tic particles,” Nature Reviews Materials, vol. 7, no. 2, pp. 138–152, 2022

    work page 2022

  5. [5]

    Your brain is full of microplastics: are they harming you?

    M. Kozlov, “Your brain is full of microplastics: are they harming you?” Nature, vol. 638, no. 8050, pp. 311–313, 2025

    work page 2025

  6. [6]

    Urban river water level increase through plastic waste accumulation at a rack structure,

    D. Honingh, T. Van Emmerik, W. Uijttewaal, H. Kardhana, O. Hoes, and N. Van de Giesen, “Urban river water level increase through plastic waste accumulation at a rack structure,”Frontiers in earth science, vol. 8, p. 28, 2020

    work page 2020

  7. [7]

    River cleaning plastic,

    “River cleaning plastic,” https://rivercleaning.com/

    work page

  8. [8]

    Blue barriers by seads: Sea defence solutions,

    “Blue barriers by seads: Sea defence solutions,” https://www. seadefencesolutions.com/

    work page

Show all 37 references
  1. [9]

    Watergoat,

    “Watergoat,” https://www.watergoat.org/

  2. [10]

    Interceptor barrier by the oceancleanup,

    “Interceptor barrier by the oceancleanup,” https://theoceancleanup.com/ rivers/

  3. [11]

    Circleaner by noria sustainable innovators,

    “Circleaner by noria sustainable innovators,” https://www.noria.earth/

  4. [12]

    Optimal design of a floating waste-collecting robot utilizing vortex phenomena,

    J. Lee, S. Roh, J. Im, M. Kim, T. Kim, and S. Yoo, “Optimal design of a floating waste-collecting robot utilizing vortex phenomena,” IEEE Access, 2024

    2024

  5. [13]

    Seabin by the seabin foundation,

    “Seabin by the seabin foundation,” https://seabinfoundation.org/

  6. [14]

    The great bubble barrier,

    “The great bubble barrier,” https://thegreatbubblebarrier.com/

  7. [15]

    Research and development of recycling boats for floating garbage,

    C. Ruixi, C. Haoming, L. Kexin, L. Xintong, X. Qiao, and Y . Jinhan, “Research and development of recycling boats for floating garbage,” in 2023 IEEE 3rd International Conference on Information Technology, Big Data and Artificial Intelligence (ICIBA) , vol. 3, 2023, pp. 1229–1235

    2023

  8. [16]

    Smurf: A fully autonomous water surface cleaning robot with a novel coverage path planning method,

    J. Zhu, Y . Yang, and Y . Cheng, “Smurf: A fully autonomous water surface cleaning robot with a novel coverage path planning method,” Journal of Marine Science and Engineering , vol. 10, no. 11, p. 1620, 2022

    2022

  9. [17]

    “Seavax,” https://www.oceansplasticleanup.com/SeaVax RiverVax/ SeaVax RiverVax Projects Overview.htm

  10. [18]

    the wasteshark by ranmarine,

    “the wasteshark by ranmarine,” https://www.ranmarine.io/products/ wasteshark/

  11. [19]

    Jellyfishbot by iadys,

    “Jellyfishbot by iadys,” https://www.jellyfishbot.io/ maritime-river-lake-water-sites/

  12. [20]

    A spiral-propulsion amphibious intelligent robot for land garbage cleaning and sea garbage cleaning,

    Y . Zhang, Z. Huang, C. Chen, X. Wu, S. Xie, H. Zhou, Y . Gou, L. Gu, and M. Ma, “A spiral-propulsion amphibious intelligent robot for land garbage cleaning and sea garbage cleaning,” Journal of Marine Science and Engineering, vol. 11, no. 8, p. 1482, 2023

    2023

  13. [21]

    A novel method for design and development of hybrid land and water buoyancy trash collecting robot,

    S. Muthusamy, S. Duraisamy, M. Ramachandran, J. Karthikeyan, J. I. David, H. K. Settu, and A. Rathinasamy, “A novel method for design and development of hybrid land and water buoyancy trash collecting robot,” in 2024 International Conference on Intelligent and Innovative Techn...

    2024

  14. [22]

    Riverine plastic litter monitoring using unmanned aerial vehicles (uavs),

    M. Geraeds, T. van Emmerik, R. de Vries, and M. S. bin Ab Razak, “Riverine plastic litter monitoring using unmanned aerial vehicles (uavs),” Remote Sensing, vol. 11, no. 17, p. 2045, 2019

    2019

  15. [23]

    Towards instance segmentation-based litter collection with multi-rotor aerial ve- hicle,

    F. Zoric, A. Franchi, M. Orsag, Z. Kovacic, and C. Gabellieri, “Towards instance segmentation-based litter collection with multi-rotor aerial ve- hicle,” in 2024 International Conference on Unmanned Aircraft Systems (ICUAS). IEEE, 2024, pp. 631–637

    2024

  16. [24]

    Differential flatness and manipulation of elasto-flexible cables carried by aerial robots in a possibly viscous environment,

    C. Gabellieri and A. Franchi, “Differential flatness and manipulation of elasto-flexible cables carried by aerial robots in a possibly viscous environment,” in 2023 International Conference on Unmanned Aircraft Systems (ICUAS). IEEE, 2023, pp. 963–968

    2023

  17. [25]

    Towards a heterogeneous cable-connected team of uavs for aerial manipulation,

    V . Abhishek, V . Srivastava, and R. Mukherjee, “Towards a heterogeneous cable-connected team of uavs for aerial manipulation,” in2021 American Control Conference. IEEE, 2021, pp. 54–59

    2021

  18. [26]

    Multiple quadrotors carrying a flexible hose: dynamics, differential flatness and control,

    P. Kotaru and K. Sreenath, “Multiple quadrotors carrying a flexible hose: dynamics, differential flatness and control,”IFAC-PapersOnLine, vol. 53, no. 2, pp. 8832–8839, 2020

    2020

  19. [27]

    The catenary robot: Design and control of a cable propelled by two quadrotors,

    D. S. D’antonio, G. A. Cardona, and D. Saldana, “The catenary robot: Design and control of a cable propelled by two quadrotors,” IEEE Robotics and Automation Letters , vol. 6, no. 2, pp. 3857–3863, 2021

    2021

  20. [28]

    Obstacle avoidance flight and shape estimation using catenary curve for manipulation of a cable hanged by aerial robots,

    S. Abiko, A. Kuno, S. Narasaki, A. Oosedo, S. Kokubun, and M. Uchiyama, “Obstacle avoidance flight and shape estimation using catenary curve for manipulation of a cable hanged by aerial robots,” in 2017 IEEE International Conference on Robotics and Biomimetics (ROBIO). IEEE, 2...

    2017

  21. [29]

    Hybrid mod- eling of deformable linear objects for their cooperative transportation by teams of quadrotors,

    J. Estevez, J. M. Lopez-Guede, G. Garate, and M. Gra ˜na, “Hybrid mod- eling of deformable linear objects for their cooperative transportation by teams of quadrotors,” Applied Sciences, vol. 12, no. 10, p. 5253, 2022

    2022

  22. [30]

    Shape visual servoing of a tether cable from parabolic features,

    L. Smolentsev, A. Krupa, and F. Chaumette, “Shape visual servoing of a tether cable from parabolic features,” in 2023 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2023, pp. 734– 740

    2023

  23. [31]

    Shape visual servoing of a cable suspended between two drones,

    ——, “Shape visual servoing of a cable suspended between two drones,” IEEE Robotics and Automation Letters , 2024

    2024

  24. [32]

    Wasted shores: Using drones to monitor the spatio-temporal evolution of debris accumulation hotspots on south africa’s umgeni river,

    T. Gutsa, C. Trois, R. de Vries, and T. Mani, “Wasted shores: Using drones to monitor the spatio-temporal evolution of debris accumulation hotspots on south africa’s umgeni river,” Science of the Total Environ- ment, vol. 955, p. 176791, 2024

    2024

  25. [33]

    Monitoring macroplastics in aquatic and terrestrial ecosystems: Expert survey reveals visual and drone-based census as most effective techniques,

    L. Gallitelli, P. Girard, U. Andriolo, M. Liro, G. Suaria, C. Martin, A. Lusher, K. Hancke, M. Blettler, O. Garcia-Garin et al., “Monitoring macroplastics in aquatic and terrestrial ecosystems: Expert survey reveals visual and drone-based census as most effective techniques,” ...

    2024

  26. [34]

    Survey on aerial manipulator: System, modeling, and control,

    X. Meng, Y . He, and J. Han, “Survey on aerial manipulator: System, modeling, and control,” Robotica, vol. 38, no. 7, pp. 1288–1317, 2020

    2020

  27. [35]

    Past, present, and future of aerial robotic manipulators,

    A. Ollero, M. Tognon, A. Suarez, D. Lee, and A. Franchi, “Past, present, and future of aerial robotic manipulators,” IEEE Transactions on Robotics, vol. 38, no. 1, pp. 626–645, 2021

    2021

  28. [36]

    Polynomial trajectory planning for aggressive quadrotor flight in dense indoor environments,

    C. Richter, A. Bry, and N. Roy, “Polynomial trajectory planning for aggressive quadrotor flight in dense indoor environments,” in Robotics Research: The 16th International Symposium ISRR . Springer, 2016, pp. 649–666

    2016

  29. [37]

    CasADi – A software framework for nonlinear optimization and opti- mal control,

    J. A. E. Andersson, J. Gillis, G. Horn, J. B. Rawlings, and M. Diehl, “CasADi – A software framework for nonlinear optimization and opti- mal control,” Mathematical Programming Computation , vol. 11, no. 1, pp. 1–36, 2019

    2019

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.