Pith. sign in

REVIEW 4 major objections 6 minor 23 references

Aerial Grasping via Maximizing Delta-Arm Workspace Utilization

T0 review · 4 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A learned feasibility map unlocks the delta arm's non-convex workspace for aerial grasping, cutting execution time and enabling a 4 m grab in 1.8 s.

desk verdict A plausible, incrementally novel aerial-grasping planner whose real-world demos are the best evidence, but the headline workspace-utilization gain is computed from an analytical model rather than from the learned feasibility model actually used in planning. read the letter →

arxiv 2506.15539 v2 pith:SRVGH7DE submitted 2025-06-18 cs.RO

Haoran Chen , Weiliang Deng , Biyu Ye , Yifan Xiong , Zongliang Pan , Ximin Lyu This is my paper · ORCID
classification cs.RO
keywords aerialgraspingdeltaarmworkspaceutilizationtrajectoryoptimizationmanipulationconstraintsreversibleresidualnetworkmachinelearning
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 tries to show that the largest untapped resource in aerial grasping with a delta arm is the arm's own non-convex workspace, and that a planner can be made to use nearly all of it by replacing the standard convex-box workspace approximation with a learned feasibility-probability map. It couples quadrotor and arm trajectories in one optimization, using an MLP to judge whether a point is reachable and a reversible residual network to approximate forward kinematics, so that workspace constraints can be removed from the optimization variables. The claimed payoff is faster, more flexible grasps: a measured 363% increase in usable workspace volume, lower simulated execution times than the convex-box baseline, and real-world pickups up to 4 m away in 1.8 s. If true, the practical consequence is that small delta-arm aerial manipulators can keep their arm folded during transit and extend only near the target, improving both speed and stability.

What carries the argument

The load-bearing machinery is a six-layer MLP $F_w$ that maps a Cartesian point in the delta-arm frame to a feasibility probability $P \in [0,1]$, and whose input gradient pulls out-of-workspace waypoints back inside the reachable region. The second piece is $\Delta$ RevNet, a reversible residual network that approximates forward kinematics from normalized joint angles $\vartheta_n \in [0,1]$ to end-effector coordinates; because it is reversible, gradients stay cheap and stable. This turns the reachable end-effector coordinate $\mathbf{q}_e$ into six unconstrained optimization variables $\Xi = [\delta_1, \zeta_1, \delta_2, \zeta_2, \delta_3, \zeta_3]^T$ via $\vartheta_n = \delta_n^2/(\zeta_n^2+\delta_n^2)$, so the non-convex workspace constraint never appears explicitly during optimization. Trajectories are carried by the piecewise-polynomial MINCO representation and optimized with L-BFGS.

What would settle it

Run the released optimizer on a target and then check every commanded end-effector position against the real delta arm's inverse kinematics; if any waypoint the optimizer believes feasible cannot be reached, or if a dense grid of physical reachability tests around the classifier's 0.5-level set disagrees with the learned boundary, the central claim fails.

Watch

Extended reading notes

Core claim

$\Delta$ arms have a pronounced non-convex reachable region, and approximating it by an inscribed rectangular prism, as prior planners do, discards the upper portion of the workspace, forcing the arm into highly extended poses and slowing captures. This paper's central claim is that a coupled whole-body trajectory optimizer can instead treat the workspace as a learned classifier: an MLP maps Cartesian points to a feasibility probability, and its differentiable gradient pulls out-of-workspace waypoints back inside. A second learned model, a RevNet approximating the delta arm's forward kinematics, makes every reachable end-effector coordinate expressible by unconstrained joint-angle variables, so the non-convex workspace constraint disappears from the optimization variables altogether. With this representation the usable workspace volume rises from 551 cm$^3$ (inscribed cube) to 2553 cm$^3$, and in benchmark scenarios execution time drops relative to the convex-box baseline, with real-world experiments completing grasps on objects 4 m away in 1.8 s and end-effector positioning errors of 0.02 m and 0.04 m.

Load-bearing premise

The planner's reliability rests on the learned feasibility classifier matching the true reachable region, especially near the 0.5 threshold; a misjudged boundary or a misleading gradient would send the planned arm outside what the hardware can do.

Editorial extensions

If this is right

  • A planner can treat the whole non-convex delta workspace as usable rather than only its largest inscribed cube, so the arm no longer has to stay fully extended during flight.
  • Because workspace constraints are eliminated from the optimization variables, the same coupled whole-body optimization machinery can be applied to other manipulators whose reachable sets are non-convex.
  • Simulated pick-and-place runs converge to roughly 55% lower objective cost than the convex-box baseline, which translates into shorter execution times.
  • Real-world grasping of objects up to 4 m away in 1.8 s, with approach speeds up to 3 m/s, shows the learned workspace representation holds up outside simulation.
  • The learned feasibility probability gives the optimizer a smooth, differentiable signal for pushing trajectory points into reachable space, which is what allows the upper, previously unused portion of the workspace to be exploited safely.

Reading between the lines

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

  • The same feasibility-probability trick could be reused for other non-convex constraints in whole-body planning, such as static-obstacle corridors, if the classifier is trained on signed distance or occupancy rather than inverse-kinematics reachability.
  • The 0.5 threshold is a safety knob: raising it buys robustness against model error while giving back some of the 363% workspace gain, a trade-off the paper does not quantify.
  • Because the RevNet is differentiable and lightweight, it could be fine-tuned online from motion-capture observations of the real arm, correcting forward-kinematics bias without changing the optimizer.
  • A natural test is to compare the MLP's 0.5-level surface against a dense alpha-shape reconstruction of the true workspace; the gap between them would directly estimate how much of the claimed 363% is real reachable space versus classifier overconfidence.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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. The paper proposes a whole-body trajectory optimization framework for an aerial manipulator consisting of a quadrotor and a delta arm. The key ideas are (i) representing the delta arm's non-convex workspace by an MLP that maps Cartesian positions to feasibility probabilities, and (ii) using a RevNet to approximate the delta arm's forward kinematics so that workspace constraints on optimization variables can be eliminated via differentiable surrogate gradients. The authors formulate the planning problem as a coupled optimization over quadrotor and arm trajectories, enforce task-specific end-effector positioning and orientation constraints, and validate the framework in simulation and in two real-world grasping experiments. The central quantitative claims are a 363% increase in usable workspace volume, faster execution times compared with an inscribed-cube baseline, an approximately 55% reduction in optimization cost, and real-world grasping of an object placed 4 m away in 1.8 s.

Significance. If the central claims are substantiated, the paper offers a practical approach to a real problem: incorporating the genuinely non-convex workspace of a delta arm into trajectory optimization, rather than conservatively approximating it by a convex volume. The learning-based representation is a reasonable surrogate strategy, and the constraint-elimination idea via a differentiable FK surrogate is coherent. The real-world demonstrations give some evidence of feasibility. At the same time, the headline workspace-utilization number is currently computed from the analytical workspace, not from the learned model that the planner actually uses, and the comparison baseline is ambiguously identified and appears to include the authors' own prior work. The reported aggregate model errors do not constrain the boundary behavior on which the optimization relies. These issues are fixable and the core framework is defensible, but the evidence as presented does not yet support the strongest claims.

major comments (4)
  1. [Section IV.A] The 363% workspace-utilization claim is not computed with the model used in planning. The paper reports that the alpha-shape reconstruction (α=10) of the analytical delta workspace has volume 2553 cm3 versus 551 cm3 for the inscribed cube, but Eq. (15) penalizes S(0.5 − P_i,j), where P comes from the MLP. The volume of the MLP's 0.5 feasible set is never reported, so the reader cannot verify that the learned representation actually expands the planner's usable workspace beyond the cube. Moreover, the reported test error below 0.001% on 100,000 uniformly sampled points is not informative near the workspace boundary, which has measure zero inside the sampling box; a systematic boundary offset of a few centimeters can coexist with very low overall error and would directly change the penalty in Eq. (15). Please report the volume and boundary error of the MLP threshold set, and compare planned trajectories against those obtained with the actual MLP feasible set rather than the analytical alpha-shape volume.
  2. [Sections II.B, IV.A, V.B] The baseline method is inconsistently identified. Related Work (Section II.B) attributes the maximum inscribed rectangular prism to Cao et al. [9] and a slender rectangular prism to Deng et al. [8]; Section IV.A says previous approaches employ the inscribed cube [9]; but Section V.B states that comparisons are performed against the method from [8], which is the authors' prior work and is elsewhere described as a slender prism. If the simulation baseline is the authors' own conservative prism, the claimed execution-time advantage and the approximately 55% cost reduction may reflect a weak baseline rather than the benefit of the new representation. Please clarify which method was implemented, report its exact feasible set, and ideally compare against a stronger baseline such as a maximal inscribed polyhedron or a state-of-the-art non-convex workspace method.
  3. [Section IV.C and Eq. (20)] The constraint-elimination route depends on the RevNet forward-kinematics surrogate being accurate in both output and gradient. The paper reports only an aggregate error below 0.001% on test samples and does not report where errors occur relative to the joint-range boundary or how Jacobian errors affect optimization convergence. If the RevNet output does not lie in the true workspace, then the claimed elimination of workspace constraints is not guaranteed, and the task constraints in Section IV.B.2 may be evaluated at unreachable end-effector poses. Please provide the distribution of FK error over the joint domain, especially near θ=0° and θ=90°, and an end-to-end comparison of optimization with the analytic FK versus the RevNet surrogate.
  4. [Section V.C] The real-world experiments demonstrate feasibility but do not compare against any baseline and therefore do not support the claim that the proposed method is faster or more flexible than alternative approaches. The 4 m / 1.8 s grasping time and speeds up to 3 m/s are standalone numbers without context. Please either run the same scenarios with the baseline planner or clearly reposition these results as feasibility demonstrations rather than comparative evidence.
minor comments (6)
  1. [Fig. 2 caption] The caption contains the phrase "continuous polymathic trajectory generation"; this should be "polynomial".
  2. [Eq. (1) and surrounding text] The phrase "the moment of the inertia matrix" should be "the moment of inertia matrix".
  3. [Throughout] There are frequent spacing artifacts in "UA V" that should be "UAV", and some inline equations have inconsistent spacing around symbols.
  4. [Section V.A] Training details are sparse: dataset sizes, sampling procedures, network hyperparameters, and train/test splits are not reported, which limits reproducibility. Please provide these details or an open-source implementation.
  5. [Section IV.B.2, Eq. (13)] The orientation constraint uses the normalized thrust vector fi(tγ) as a proxy for orientation. Please clarify under what flight conditions this is a valid proxy and whether the constraint was active in the reported experiments.
  6. [Fig. 8 caption] The caption refers to "the two methods" in parts (a-b) and (c-d), but the real-world experiments appear to test only the proposed method. Please reconcile the caption with the content.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the learned surrogates are fitted to the analytic delta-arm kinematics and the performance claims are tested externally; the workspace-volume figure is a geometric property, not a fitted prediction.

full rationale

The paper's derivation chain is not circular. The MLP feasibility classifier (Sec. IV-A) is trained on inverse-kinematics labels for the same delta-arm workspace used later in Eq. (15); that makes it a supervised approximator of the indicator function of the workspace rather than an independent source of the workspace geometry. The RevNet forward-kinematics model (Sec. IV-C) is likewise trained on forward-kinematics samples and is used to reparameterize the end-effector intermediate points by unconstrained joint-angle variables, so the feasibility of those points is enforced by construction through the joint-angle range. This is a surrogate-modeling design, not a case where the output is defined to be the input. The headline 363% workspace-utilization figure is computed by reconstructing the analytic delta workspace with the alpha-shape method and comparing its volume with the inscribed cube (Sec. IV-A). Although the paper attributes this volume to 'our method,' the volume does not come from the MLP's 0.5 level set; this is an attribution/validation gap rather than a circular reduction, since the planner's final parameterization indeed ranges over the full joint-angle cube. The comparison baseline [8] and the controller [20] are self-citations, but they are used as benchmarks and implementation references, not as the justification for the paper's central claim. No equation in the paper is shown to equal its own input by construction, and no fitted parameter is renamed as a prediction. Under the stated criteria, there is no significant circularity.

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

The central claim depends on learned models trained on the arm's kinematic model rather than on independently verified physical data, and on standard trajectory optimization components taken from prior work. No new physical entities are introduced.

free parameters (5)
  • workspace feasibility threshold = 0.5
    Hand-chosen threshold in Eq. (15) to classify points as inside or outside the delta workspace. The effective workspace volume and optimizer behavior depend on this threshold.
  • alpha shape parameter alpha = 10
    Used in the MATLAB reconstruction of the delta workspace to compute the reported volume 2553 cm3. The volume comparison depends on this reconstruction parameter.
  • MLP architecture hyperparameters = 64-256-128-64 layers
    Layer widths chosen by hand for the workspace representation model. These affect approximation accuracy and gradient quality, which underpin the central feasibility-constraint mechanism.
  • RevNet architecture hyperparameters = four RevBlocks, hidden sizes not fully specified
    The structure of the learned forward kinematics model; the paper claims error below 0.001% but exact architecture details are incomplete.
  • constraint penalty weights w_u = not reported
    The objective (Eq. 9) includes weights for each constraint penalty. Their values affect the trajectory quality and are not disclosed.
assumptions (5)
  • domain assumption The simplified aerial manipulator dynamics and the delta-arm forward kinematics model from [20] accurately represent the physical platform.
    The dynamic equations (Eq. 1) and the forward kinematics h_e(theta) are taken from prior work and used to generate training data and simulate the system.
  • standard math The MINCO trajectory parameterization (Theorem 2 in [21]) correctly maps intermediate points q and time T to polynomial coefficients for the given 6D trajectory.
    The trajectory representation in Section III.B relies on this theorem from the cited literature.
  • domain assumption The training samples for the MLP and RevNet are representative of the true workspace and forward kinematics, so the learned models generalize to unseen inputs.
    The models are trained on randomly sampled points, but the paper does not specify the number of samples or the sampling distribution, so representativeness is assumed.
  • ad hoc to paper The learned feasibility probability with a 0.5 threshold accurately corresponds to the true reachable workspace boundary.
    The workspace penalty (Eq. 15) uses 0.5 as a hard threshold, but no analysis is given of the probability calibration near the boundary.
  • domain assumption Encoding the grasping orientation constraint via the normalized thrust vector f_i (Eq. 13) is sufficient to achieve the desired end-effector orientation.
    The task constraint assumes that aligning the thrust direction with the target orientation ensures correct grasp approach, which may be an oversimplification.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Aerial Grasping via Maximizing Delta-Arm Workspace Utilization." pith.science (2026). https://pith.science/paper/SRVGH7DE

@misc{pith2026250615539,
  author       = {Pith},
  title        = {Pith review of: Aerial Grasping via Maximizing Delta-Arm Workspace Utilization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SRVGH7DE}},
  note         = {Machine review of arXiv:2506.15539}
}
read the original abstract

The workspace limits the operational capabilities and range of motion for the systems with robotic arms. Maximizing workspace utilization has the potential to provide more optimal solutions for aerial manipulation tasks, increasing the system's flexibility and operational efficiency. In this paper, we introduce a novel planning framework for aerial grasping that maximizes workspace utilization. We formulate an optimization problem to optimize the aerial manipulator's trajectory, incorporating task constraints to achieve efficient manipulation. To address the challenge of incorporating the delta arm's non-convex workspace into optimization constraints, we leverage a Multilayer Perceptron (MLP) to map position points to feasibility probabilities.Furthermore, we employ Reversible Residual Networks (RevNet) to approximate the complex forward kinematics of the delta arm, utilizing efficient model gradients to eliminate workspace constraints. We validate our methods in simulations and real-world experiments to demonstrate their effectiveness.

Figures

Figures reproduced from arXiv: 2506.15539 by the authors.

Figure 1
Figure 1. The aerial manipulator successfully navigates to the target [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overview of our proposed manipulation planning framework for aerial manipulators with delta arms, which mainly consists of [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Workspace representation method. (a) Comparison of base￾line representation method [9] and ours. (b) The workflow of our representation model. (c) The process of how an optimization point is navigated to the workspace [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Model structure. (a) Workspace representation model using MLP. (b) Forward kinematics model using RevNet. the optimization process, this gradient guides points initially located outside the workspace towards higher-probability feasible regions as shown in [PITH_FULL_I…
Figure 5
Figure 5. Figure 5: Comparison of the two methods in Scenario 1. [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 7
Figure 7. Figure 7: Snapshot of aerial grasping in real-world scenarios with [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 8
Figure 8. Figure 8: Flight data from real-world experiments. [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

23 extracted references · 19 canonical work pages

  1. [9]

    Motion planning for aerial pick-and-place with geometric feasibility constraints,

    H. Cao, J. Shen, C. Liu, B. Zhu, and S. Zhao, “Motion planning for aerial pick-and-place with geometric feasibility constraints,” IEEE Transactions on Automation Science and Engineering , vol. 22, pp. 2577–2594, 2025

    work page 2025

  2. [8]

    Whole-body integrated motion planning for aerial manipulators,

    W. Deng, H. Chen, B. Ye, H. Chen, and X. Lyu, “Whole-body integrated motion planning for aerial manipulators,” arXiv preprint arXiv:2501.06493, 2025

    arXiv 2025

  3. [1]

    Autonomous marker-less rapid aerial grasping,

    E. Bauer, B. G. Cangan, and R. K. Katzschmann, “Autonomous marker-less rapid aerial grasping,” in 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2023, pp. 6395–6402

    work page 2023

  4. [2]

    Aerial additive manufacturing with multiple autonomous robots,

    K. Zhang, P. Chermprayong, F. Xiao, D. Tzoumanikas, B. Dams, S. Kay, B. Kocer, A. Burns, L. Orr, T. Alhinai, C. Choi, D. Darekar, W. Li, S. Hirschmann, V . Soana, S. Ngah, C. Grillot, S. Sareh, A. Choubey, and M. Kovac, “Aerial additive manufacturing with multiple autonomous robots,” Nature, vol. 609, pp. 709–717, 09 2022

    work page 2022

  5. [3]

    Perching and resting—a paradigm for uav maneuver- ing with modularized landing gears,

    K. Hang, L. Ximin, H. Song, J. Stork, A. Dollar, D. Kragic, and F. Zhang, “Perching and resting—a paradigm for uav maneuver- ing with modularized landing gears,” Science Robotics , vol. 4, p. eaau6637, 03 2019

    work page 2019

  6. [4]

    Thrust microstepping via acceleration feedback in quadrotor control for aerial grasping of dynamic payload,

    A. Kumar and L. Behera, “Thrust microstepping via acceleration feedback in quadrotor control for aerial grasping of dynamic payload,” IEEE Robotics and Automation Letters , vol. 9, no. 2, pp. 1246–1253, 2024

    work page 2024

  7. [5]

    Impact absorbing and compensation for heavy object catching with an unmanned aerial manipulator,

    S. Wang, Z. Ma, F. Quan, and H. Chen, “Impact absorbing and compensation for heavy object catching with an unmanned aerial manipulator,” IEEE Robotics and Automation Letters , vol. 9, no. 4, pp. 3656–3663, 2024-04

    work page 2024

  8. [6]

    Powerful uav manipulation via bioin- spired self-adaptive soft self-contained gripper,

    X. Guo, W. Tang, K. Qin, Y . Zhong, H. Xu, Y . Qu, Z. Li, Q. Sheng, Y . Gao, H. Yang, and J. Zou, “Powerful uav manipulation via bioin- spired self-adaptive soft self-contained gripper,” Science Advances , vol. 10, no. 19, p. eadn6642, 2024

    work page 2024

Show all 23 references
  1. [7]

    Perching and grasping using a passive dynamic bioinspired gripper,

    A. Firouzeh, J. Lee, H. Yang, D. Lee, and K.-J. Cho, “Perching and grasping using a passive dynamic bioinspired gripper,” IEEE Transactions on Robotics , vol. 40, pp. 213–225, 2024

    2024

  2. [10]

    A small form factor aerial research vehicle for pick- and-place tasks with onboard real-time object detection and visual odometry,

    C. A. Dimmig, A. Goodridge, G. Baraban, P. Zhu, J. Bhowmick, and M. Kobilarov, “A small form factor aerial research vehicle for pick- and-place tasks with onboard real-time object detection and visual odometry,” in 2023 IEEE/RSJ International Conference on Intelligent Robots a...

    2023

  3. [11]

    Improving the reliability of pick-and-place with aerial vehicles through fault-tolerant software and a custom magnetic end-effector,

    G. Garimella, M. Sheckells, S. Kim, G. Baraban, and M. Kobi- larov, “Improving the reliability of pick-and-place with aerial vehicles through fault-tolerant software and a custom magnetic end-effector,” IEEE Robotics and Automation Letters , vol. 6, no. 4, pp. 7501–7508, 2021

    2021

  4. [12]

    Aerial grasping with a lightweight manipulator based on multi-objective optimization and visual compensation,

    H. Chen, F. Quan, L. Fang, and S. Zhang, “Aerial grasping with a lightweight manipulator based on multi-objective optimization and visual compensation,” Sensors, vol. 19, no. 19, p. 4253, 2019

    2019

  5. [13]

    Grasp planning and visual servoing for an outdoors aerial dual manipulator,

    P. Ramon, B. Arrue, and A. Baturone, “Grasp planning and visual servoing for an outdoors aerial dual manipulator,” Engineering, vol. 6, 11 2019

    2019

  6. [14]

    Sequential manipulation planning for over-actuated unmanned aerial manipulators,

    Y . Su, J. Li, Z. Jiao, M. Wang, C. Chu, H. Li, Y . Zhu, and H. Liu, “Sequential manipulation planning for over-actuated unmanned aerial manipulators,” in 2023 IEEE/RSJ International Conference on Intelli- gent Robots and Systems (IROS) . IEEE, 2023, pp. 6905–6911

    2023

  7. [15]

    Planning and control for collision-free cooperative aerial transportation,

    H. Lee, H. Kim, and H. J. Kim, “Planning and control for collision-free cooperative aerial transportation,” IEEE Transactions on Automation Science and Engineering , vol. 15, no. 1, pp. 189–201, 2018

    2018

  8. [16]

    Control-aware motion planning for task-constrained aerial manipulation,

    M. Tognon, E. Cataldi, H. A. T. Chavez, G. Antonelli, J. Cortes, and A. Franchi, “Control-aware motion planning for task-constrained aerial manipulation,” IEEE Robotics and Automation Letters , vol. 3, no. 3, pp. 2478–2484, 2018

    2018

  9. [17]

    Bubble planner: Planning high-speed smooth quadrotor trajectories using receding corridors,

    Y . Ren, F. Zhu, W. Liu, Z. Wang, Y . Lin, F. Gao, and F. Zhang, “Bubble planner: Planning high-speed smooth quadrotor trajectories using receding corridors,” in 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2022, pp. 6332– 6339

    2022

  10. [18]

    Learning to open doors with an aerial manipulator,

    E. Cuniato, I. Geles, W. Zhang, O. Andersson, M. Tognon, and R. Siegwart, “Learning to open doors with an aerial manipulator,” in 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2023, pp. 6942–6948

    2023

  11. [19]

    Capturing robot workspace structure: Representing robot capabilities,

    F. Zacharias, C. Borst, and G. Hirzinger, “Capturing robot workspace structure: Representing robot capabilities,” in 2007 IEEE/RSJ Interna- tional Conference on Intelligent Robots and Systems . IEEE, 2007, pp. 3229–3236

    2007

  12. [20]

    Ndob-based control of a uav with delta-arm considering manipulator dynamics,

    H. Chen, B. Ye, X. Liang, W. Deng, and X. Lyu, “Ndob-based control of a uav with delta-arm considering manipulator dynamics,” arXiv preprint arXiv:2501.06122, 2025

    2025

  13. [21]

    Geometrically constrained tra- jectory optimization for multicopters,

    Z. Wang, X. Zhou, C. Xu, and F. Gao, “Geometrically constrained tra- jectory optimization for multicopters,” IEEE Transactions on Robotics, vol. 38, no. 5, pp. 3259–3278, 2022

    2022

  14. [22]

    Neural network approxima- tion,

    R. DeV ore, B. Hanin, and G. Petrova, “Neural network approxima- tion,” Acta Numerica, vol. 30, pp. 327–444, 2021

    2021

  15. [23]

    The reversible residual network: Backpropagation without storing activations,

    A. N. Gomez, M. Ren, R. Urtasun, and R. B. Grosse, “The reversible residual network: Backpropagation without storing activations,” Ad- vances in neural information processing systems , vol. 30, 2017

    2017

Pith tools

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