Pith. sign in

REVIEW 4 major objections 4 minor 33 references

Automatic Generation of Aerobatic Flight in Complex Environments via Diffusion Models

T0 review · 4 major / 4 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read A diffusion model turns short aerobatic clips into long, collision-free drone flights.

desk verdict Credible aerobatic-primitive diffusion pipeline with informative ablations, but the final collision-free claim is unsupported because the post-processing safety constraint is only an integral inequality. read the letter →

arxiv 2504.15138 v1 pith:M6AMBJVY submitted 2025-04-21 cs.RO

Yuhang Zhong , Anke Zhao , Tianyue Wu , Tingrui Zhang , Fei Gao This is my paper · ORCID
classification cs.RO
keywords diffusionmodelsaerobaticflightquadrotortrajectorygenerationcollisionavoidanceclassifierguidanceoptimizationprimitivesmotion
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

This paper claims that long-horizon aerobatic drone flight can be generated automatically by learning short, reusable maneuver segments called aerobatic primitives and chaining them with a conditional diffusion model. The model is trained only on obstacle-free, optimization-generated demonstrations, yet at inference time per-step signed-distance-field guidance, batch sampling, and a coarse collision check steer the output into collision-free trajectories in unseen cluttered environments. A hierarchical spatial-temporal trajectory optimization then turns the discrete generated frames into dynamically feasible commands that a real quadrotor can track. If the claim holds, non-experts could specify target waypoints and maneuver styles and receive executable, user-editable aerobatic sequences without manual trajectory tuning.

What carries the argument

The central object is the aerobatic primitive: a short frame sequence (position plus continuous 6-DoF rotation, plus a padding flag) that captures the attitude dynamics of one maneuver and can be chained with others. The paper couples this representation with a decoder-only Diffusion Transformer that jointly attends to the noisy current primitive and five frames of historical observations, enforcing continuity between successive primitives while conditioning on target waypoint, maneuver-style action, and denoising time step. Two mechanisms carry the argument: classifier guidance using an SDF-based collision cost to shift denoised samples away from obstacles, and batch sampling with a coarse collision check to rescue colliding candidates. The final load-bearing component is hierarchical trajectory optimization on the MINCO representation, which first solves a relaxed problem without z-axis angular-velocity constraints and then refines with full dynamic constraints, producing commands a low-level controller can track.

What would settle it

Run the trained model in an environment with corridors narrower than the 1 m waypoint resolution used in training, or with obstacle gradients that conflict with the learned maneuver geometry, and measure the collision-free success rate over many seeds; a sharp drop as obstacle density approaches the primitive scale would show the guidance does not generalize beyond the tested scenes.

Watch

Extended reading notes

Core claim

The paper's central discovery is that a diffusion model trained exclusively on short-horizon aerobatic primitives in open space can generate arbitrary long-horizon sequences of aerobatic maneuvers in complex environments, provided each primitive is conditioned on historical state observations and the sampling process is steered online by a collision cost. The authors decompose maneuvers such as Power Loop, Barrel Roll, Split-S, Immelmann Turn, and Wall Ride into short state sequences, append redundant segments to model transitions, and train a Diffusion Transformer to denoise a fixed-length frame window that is truncated by a state flag. At inference, classifier guidance built from a precomputed signed distance field shifts the denoising mean away from obstacles, batch sampling produces many candidates, and a coarse collision check replaces colliding samples with collision-free alternatives. The generated dense frames are then distilled into sparse keyframe waypoints and a safe flight corridor, and a hierarchical MINCO-based trajectory optimization recovers a dynamically feasible trajectory. Real-world flights of five successive maneuvers in a 12x6x4 m indoor space with tracking errors below 0.15 m and 15 degrees support the claim.

Load-bearing premise

The entire collision-avoidance strategy depends on the assumption that per-step signed-distance-field guidance and batch resampling can redirect maneuvers learned in open space into safe trajectories in unseen cluttered environments without corrupting the aerobatic style; if that steering fails, the long-horizon claim collapses.

Editorial extensions

If this is right

  • A user can specify a sequence of target waypoints and optional maneuver-style commands and receive a long, collision-free, dynamically feasible aerobatic trajectory without manually tuning parameters.
  • Because the model trains only on short-horizon, obstacle-free demonstrations, the method removes the need for scarce long-horizon expert data in other agile-maneuver domains.
  • The post-processing stage implies that the raw diffusion output is a topological suggestion rather than an executable plan; dynamic feasibility is guaranteed only after the hierarchical optimization.
  • The ablation results indicate that cost guidance, not the learned prior, contributes most of the collision-avoidance success, so safety scales with the quality of the environment map and the batch size.

Reading between the lines

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

  • The same primitive-composition recipe could transfer to other vehicles with discrete maneuver vocabularies, such as fixed-wing tailsitters or ground robots, but the paper does not test that transfer.
  • Because the reported success rates stay near 97-100 percent only with a batch size of 500, a natural stress test is to measure how success degrades as batch size shrinks; this would expose how much of the safety margin comes from computation rather than from the learned model.
  • The paper's stated future direction, scene-aware generation that exploits environmental features such as flips through narrow gaps, would require the model to see obstacle geometry during training, which the current design deliberately avoids.
  • A reader should expect the method to inherit the map's accuracy: the SDF guidance is only as reliable as the precomputed distance field, so dynamic or badly modeled obstacles would break the safety guarantee.
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 / 4 minor

Summary. The paper proposes AeroDM, a framework that uses a conditional diffusion model to generate aerobatic primitives (short maneuver snippets) conditioned on historical observations, target waypoints, and optional action labels. During inference, classifier guidance using an SDF-based collision cost, batch sampling, and a coarse collision check are used to steer generation around obstacles. The generated sequences are then refined by a MINCO-based spatial-temporal trajectory optimizer with a hierarchical two-stage procedure to enforce dynamic feasibility. The method is evaluated with ablations in three simulated cluttered environments and a single real-world indoor flight of five aerobatic maneuvers.

Significance. If the claims are sustained, the paper would make a useful step toward automating long-horizon aerobatic trajectory design in cluttered scenes without per-maneuver manual tuning. The primitive representation with a padding flag and historical-conditioned generation is a sensible response to the variable-length and continuity challenges, and the ablation studies provide evidence that target/action conditioning, historical context, cost guidance, and post-processing each contribute to the reported behavior. The real-world demonstration is a genuinely nontrivial systems effort. However, the formal safety guarantee for the final executed trajectory is not established because the corridor constraint is mis-specified, and the empirical collision rates are reported on the pre-optimization outputs. These issues bear directly on the paper's central claim that the pipeline outputs long-horizon collision-free trajectories in complex environments.

major comments (4)
  1. [§IV, Eq. (15)] The safety constraint G_safe = ∫(A_i p(t) − b_i) dt ≤ 0 is an integral inequality, not a pointwise containment condition. The text states that "each trajectory segment must stay inside the corresponding ith polyhedron," but the integral can be non-positive while the trajectory leaves the polyhedron and then returns, because a positive excursion outside is offset by a longer negative interior portion. Since the executed trajectory is the post-processed MINCO trajectory, this formal gap means the collision-free property of the final output is unsupported. Please enforce pointwise membership (e.g., at collocation points or via a conservative inner approximation) or, failing that, report a post-optimization pointwise collision-check statistic for all experiments.
  2. [§V-B, Table I] The success rates in Table I are measured on the generated (pre-optimization) trajectories, as the text says "collision checks are performed on both individual motion frames and interpolated trajectories between consecutive frames." The trajectory optimizer can modify the path and timing, and with the flawed constraint in Eq. (15) there is no guarantee that the optimized trajectory inherits collision-freeness. Therefore, the high success rates do not substantiate the claim that the final pipeline produces collision-free trajectories in complex environments. Please report collision success on the final optimized outputs for each environment and for each Naero, including the real-world case.
  3. [Contribution 1 and §V-B] The claim of "arbitrary long-horizon trajectories" is not supported by the evidence. Table I goes only up to Naero=10, and the real-world experiment uses five maneuvers. Moreover, the success rates decline with Naero (e.g., Outdoor Factory drops to 97.2 ± 2.8 at Naero=10; UnCheck drops to 7.0 ± 1.6), so no evidence is presented that the composition remains reliable for much longer horizons. Please either soften the claim to the tested horizon range or add longer-horizon experiments that demonstrate the scaling behavior.
  4. [§III-B, §III-E, §IV] The model is trained on environment-agnostic demonstrations produced by an optimization pipeline, and the post-processing uses MINCO-based optimization from the same research lineage. This means the method inherits the biases and failure modes of that optimizer, and the evaluation does not quantify how the pipeline behaves in environments with obstacle layouts that are structurally different from the presented factory, workshop, and forest scenes. This is a correctness-risk concern rather than an internal inconsistency; a concrete test would be to evaluate in randomly generated maze-like environments or with systematically varied obstacle density to bound the generalization envelope.
minor comments (4)
  1. [Table I] In the Indoor Workshop row for Naero=3, the entry "26 .± 16.5" appears to have a missing digit; please fix the typesetting.
  2. [Fig. 5] The caption uses "F-OODS" and "N-OODS" while the text refers to "F-OODS" and "N-OODS"; also "UnCondS" in the figure legend and "UncondS" in the text are inconsistent. Please standardize these labels.
  3. [§III-A] The padding flag s is introduced to truncate variable-length primitives, but the training loss in Eq. (3) and Eq. (4) does not show how padding states are masked. Please clarify whether the reconstruction and velocity losses are applied only to non-padding states, and how the truncation is handled during inference.
  4. [§V-C] The real-world evaluation is a single favorable flight; it would strengthen the paper to report the number of repeated flights and the variance of tracking errors across trials, given the claim of deployability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the long-horizon and collision-free claims are evaluated against external metrics, and the self-cited optimization components are independent tools.

full rationale

I walked the claimed derivation chain: aerobatic primitives are generated by an offline optimization method (Sec. III-B), AeroDM is trained on those short-horizon demonstrations (Sec. III-C), inference chains primitives using classifier guidance from an SDF cost, batch sampling, and a coarse collision check (Sec. III-E), and finally MINCO-based spatial-temporal optimization converts the discrete primitives into dynamically feasible trajectories (Sec. IV). None of these steps defines the output as the input. The target waypoint is indeed the terminal position used as the conditioning variable, so the target-hit ablation partly confirms the model honors its condition; however, this is a sanity check of a conditional generator, not a prediction, and the out-of-distribution target result is an empirical generalization that could have failed. The collision-avoidance success rates are measured on generated frames and interpolated trajectories after the method's own coarse-rejection step; this makes the metric a system acceptance rate, but the paper explicitly presents the coarse check as part of the method, not as an emergent model capability or as a fitted parameter renamed as a prediction. The post-processing's kinodynamic constraints, MINCO representation, and safe-corridor generation are supported by self-citations [23], [32], [33], but these are code-reproduced, parameter-free tools whose assumptions do not include the paper's central claim of long-horizon aerobatic generation; hence they are independent evidence and do not raise the circularity score. The concern that Eq. (15) is an integral constraint rather than pointwise containment is a correctness/safety issue, not a circularity, because the reported collision checks are performed on generated trajectories rather than on the post-processed output. Overall, the central claims are not forced by construction.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central pipeline depends on standard diffusion theory, standard quadrotor differential-flatness assumptions, a hand-built expert dataset, and several unreported hyperparameters. No new physical entities are postulated.

free parameters (5)
  • guidance cost weights lambda_i
    Eq. 7 weights each SDF cost gradient; the values are not reported and directly control obstacle-avoidance strength.
  • safety distance d
    Eq. 8 adds a penalty when the SDF distance is below d; this hand-set margin changes how strongly trajectories are kept away from obstacles.
  • keyframe angle threshold alpha
    Section IV selects waypoints for optimization when the body z-axis deviates by more than alpha; alpha is not reported and controls fidelity of the optimized trajectory to the generated maneuver.
  • batch size = 500
    The coarse collision check replaces colliding samples only if the batch is large enough; no sensitivity analysis is provided.
  • diffusion denoising steps = 30
    Generation uses 30 steps with an exponential noise scheduler; no ablation is given for this choice.
assumptions (4)
  • standard math The Gaussian approximation used in classifier guidance (Eq. 6) is accurate enough to steer the denoising process toward collision-free samples.
    The paper adopts this approximation from [10] without re-derivation; it is standard in diffusion guidance but remains an approximation for the SDF cost.
  • domain assumption Quadrotor dynamics are differentially flat, so a position trajectory determines attitude, thrust, and angular velocity.
    Post-processing relies on this property through MINCO and differential flatness from [23, 33]; it is standard for quadrotors but not proven in the paper.
  • ad hoc to paper The optimization-generated expert dataset faithfully captures the diversity of aerobatic primitives and transitions.
    Section III-B describes dataset creation by an unspecified optimization-based method and predefined maneuver design rules; the coverage of this dataset is load-bearing for learning.
  • domain assumption Sequential conditioning on a 5-frame history plus redundant transition segments yields dynamically coherent concatenation of primitives.
    The model is trained on short segments, and the paper assumes that history conditioning and post-processing remove discontinuities; the ablation and real flight support this only for tested cases.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Automatic Generation of Aerobatic Flight in Complex Environments via Diffusion Models." pith.science (2026). https://pith.science/paper/M6AMBJVY

@misc{pith2026250415138,
  author       = {Pith},
  title        = {Pith review of: Automatic Generation of Aerobatic Flight in Complex Environments via Diffusion Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/M6AMBJVY}},
  note         = {Machine review of arXiv:2504.15138}
}
read the original abstract

Performing striking aerobatic flight in complex environments demands manual designs of key maneuvers in advance, which is intricate and time-consuming as the horizon of the trajectory performed becomes long. This paper presents a novel framework that leverages diffusion models to automate and scale up aerobatic trajectory generation. Our key innovation is the decomposition of complex maneuvers into aerobatic primitives, which are short frame sequences that act as building blocks, featuring critical aerobatic behaviors for tractable trajectory synthesis. The model learns aerobatic primitives using historical trajectory observations as dynamic priors to ensure motion continuity, with additional conditional inputs (target waypoints and optional action constraints) integrated to enable user-editable trajectory generation. During model inference, classifier guidance is incorporated with batch sampling to achieve obstacle avoidance. Additionally, the generated outcomes are refined through post-processing with spatial-temporal trajectory optimization to ensure dynamical feasibility. Extensive simulations and real-world experiments have validated the key component designs of our method, demonstrating its feasibility for deploying on real drones to achieve long-horizon aerobatic flight.

Figures

Figures reproduced from arXiv: 2504.15138 by the authors.

Figure 1
Figure 1. Our method enables automatic generation of successive long-horizon aerobatic maneuvers, allowing drones to traverse through a [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Illustration of aerobatic primitive generation, the trajectory [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. The architecture of the diffusion process. (A) Schematic [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Five different maneuver styles of aerobatic trajectories: (a) the Power Loop, (b) the Barrel Roll, (c) the Split-S, (d) the Immelmann [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Up: results of the aerobatic generation conditioned on the [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 7
Figure 7. Figure 7: The illustration of the drone executing aerobatic maneuvers in three different scenarios: (a) Narrow indoor industrial workshop, [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 8
Figure 8. Figure 8: Snapshot of a quadrotor executing aerobatic flight trajecto [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 24 canonical work pages

  1. [1]

    Let’s fly! an analysis of flying fpv drones through an online survey.,

    D. Tezza, D. Caprio, D. Laesker, and M. Andujar, “Let’s fly! an analysis of flying fpv drones through an online survey.,” in iHDI@ CHI, 2020

    work page 2020

  2. [2]

    Deep drone acrobatics,

    E. Kaufmann, A. Loquercio, R. Ranftl, M. M ¨uller, V . Koltun, and D. Scaramuzza, “Deep drone acrobatics,” ArXiv, vol. abs/2006.05768, 2020

    arXiv 2006

  3. [3]

    Trajectory generation and tracking control for aggressive tail-sitter flights,

    G. Lu, Y . Cai, N. Chen, F. Kong, Y . Ren, and F. Zhang, “Trajectory generation and tracking control for aggressive tail-sitter flights,” The International Journal of Robotics Research , vol. 43, pp. 241 – 280, 2022

    work page 2022

  4. [4]

    On-manifold model predictive control for trajectory tracking on robotic systems,

    G. Lu, W. Xu, and F. Zhang, “On-manifold model predictive control for trajectory tracking on robotic systems,” IEEE Transactions on Industrial Electronics, vol. 70, pp. 9192–9202, 2023

    work page 2023

  5. [5]

    Aerobatic trajectory generation for a vtol fixed-wing aircraft using differential flatness,

    E. Tal, G. Ryou, and S. Karaman, “Aerobatic trajectory generation for a vtol fixed-wing aircraft using differential flatness,” IEEE Transactions on Robotics , vol. 39, pp. 4805–4819, 2022

    work page 2022

  6. [6]

    Robot motion diffusion model: Motion generation for robotic characters,

    A. Serifi, E. Z ¨urich, S. D. Research, D. R. S. E. K. R. GRANDIA, E. Z. S. M. GROSS, and S. M. B ¨acher, “Robot motion diffusion model: Motion generation for robotic characters,” in ACM SIGGRAPH Conference and Exhibition on Computer Graphics and Interactive Techniques in Asia , 2024

    work page 2024

  7. [7]

    Gener- ating human interaction motions in scenes with text control,

    H. Yi, J. Thies, M. J. Black, X. B. Peng, and D. Rempe, “Gener- ating human interaction motions in scenes with text control,” ArXiv, vol. abs/2404.10685, 2024

    arXiv 2024

  8. [8]

    Flexible motion in-betweening with diffusion models,

    S. Cohan, G. Tevet, D. Reda, X. B. Peng, and M. van de Panne, “Flexible motion in-betweening with diffusion models,” in Interna- tional Conference on Computer Graphics and Interactive Techniques , 2024

    work page 2024

Show all 33 references
  1. [9]

    Planning with diffusion for flexible behavior synthesis,

    M. Janner, Y . Du, J. B. Tenenbaum, and S. Levine, “Planning with diffusion for flexible behavior synthesis,” in International Conference on Machine Learning , 2022

    2022

  2. [10]

    Motion planning diffusion: Learning and planning of robot motions with diffu- sion models,

    J. Carvalho, A. T. Le, M. Baierl, D. Koert, and J. Peters, “Motion planning diffusion: Learning and planning of robot motions with diffu- sion models,” 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , pp. 1916–1923, 2023

    2023

  3. [11]

    Diffusion-based generation, optimization, and planning in 3d scenes,

    S. Huang, Z. Wang, P. Li, B. Jia, T. Liu, Y . Zhu, W. Liang, and S.-C. Zhu, “Diffusion-based generation, optimization, and planning in 3d scenes,” 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pp. 16750–16761, 2023

    2023

  4. [12]

    Edge: Editable dance generation from music,

    J.-H. Tseng, R. Castellon, and C. K. Liu, “Edge: Editable dance generation from music,” 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pp. 448–458, 2022

    2023

  5. [13]

    Generative skill chaining: Long-horizon skill planning with diffusion models,

    U. A. Mishra, S. Xue, Y . Chen, and D. Xu, “Generative skill chaining: Long-horizon skill planning with diffusion models,” in Conference on Robot Learning , 2023

    2023

  6. [14]

    Longdancediff: Long-term dance generation with conditional diffusion model,

    S. Yang, Z. Yang, and Z. Wang, “Longdancediff: Long-term dance generation with conditional diffusion model,” ArXiv, vol. abs/2308.11945, 2023

    2023 arXiv

  7. [15]

    Closd: Closing the loop between simulation and diffusion for multi-task character control,

    G. Tevet, S. Raab, S. Cohan, D. Reda, Z. Luo, X. B. Peng, A. Bermano, and M. van de Panne, “Closd: Closing the loop between simulation and diffusion for multi-task character control,” ArXiv, vol. abs/2410.03441, 2024

    2024 arXiv

  8. [16]

    Repaint: Inpainting using denoising diffusion probabilistic models,

    A. Lugmayr, M. Danelljan, A. Romero, F. Yu, R. Timofte, and L. V . Gool, “Repaint: Inpainting using denoising diffusion probabilistic models,” 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pp. 11451–11461, 2022

    2022

  9. [17]

    Diffusion models beat gans on image synthesis,

    P. Dhariwal and A. Nichol, “Diffusion models beat gans on image synthesis,” ArXiv, vol. abs/2105.05233, 2021

    2021 arXiv

  10. [18]

    Score-based generative modeling through stochastic differential equations,

    Y . Song, J. N. Sohl-Dickstein, D. P. Kingma, A. Kumar, S. Ermon, and B. Poole, “Score-based generative modeling through stochastic differential equations,” ArXiv, vol. abs/2011.13456, 2020

    2011 arXiv

  11. [19]

    Diffusion policy: Visuomotor policy learning via action diffusion,

    C. Chi, Z. Xu, S. Feng, E. Cousineau, Y . Du, B. Burchfiel, R. Tedrake, and S. Song, “Diffusion policy: Visuomotor policy learning via action diffusion,” The International Journal of Robotics Research , 2024

    2024

  12. [20]

    Nomad: Goal masked diffusion policies for navigation and exploration,

    A. K. Sridhar, D. Shah, C. Glossop, and S. Levine, “Nomad: Goal masked diffusion policies for navigation and exploration,” 2024 IEEE International Conference on Robotics and Automation (ICRA) , pp. 63– 70, 2023

    2024

  13. [21]

    Minimum snap trajectory generation and control for quadrotors,

    D. Mellinger and V . R. Kumar, “Minimum snap trajectory generation and control for quadrotors,” 2011 IEEE International Conference on Robotics and Automation , pp. 2520–2525, 2011

    2011

  14. [22]

    Differential flatness of quadrotor dynamics subject to rotor drag for accurate tracking of high-speed trajectories,

    M. Faessler, A. Franchi, and D. Scaramuzza, “Differential flatness of quadrotor dynamics subject to rotor drag for accurate tracking of high-speed trajectories,” IEEE Robotics and Automation Letters, vol. 3, pp. 620–626, 2017

    2017

  15. [23]

    Robust trajectory planning for spatial- temporal multi-drone coordination in large scenes,

    Z. Wang, C. Xu, and F. Gao, “Robust trajectory planning for spatial- temporal multi-drone coordination in large scenes,” 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , pp. 12182–12188, 2021

    2022

  16. [24]

    Controller synthesis and perfor- mance optimization for aerobatic quadrotor flight,

    Y . Chen and N. O. P ´erez-Arancibia, “Controller synthesis and perfor- mance optimization for aerobatic quadrotor flight,” IEEE Transactions on Control Systems Technology , vol. 28, pp. 2204–2219, 2020

    2020

  17. [25]

    A simple learning strategy for high-speed quadrocopter multi-flips,

    S. Lupashin, A. P. Schoellig, M. Sherback, and R. D’Andrea, “A simple learning strategy for high-speed quadrocopter multi-flips,” 2010 IEEE International Conference on Robotics and Automation , pp. 1642–1648, 2010

    2010

  18. [26]

    Swarm of micro flying robots in the wild,

    X. Zhou, X. Wen, Z. Wang, Y . Gao, H. Li, Q. Wang, T. Yang, H. Lu, Y . Cao, C. Xu, and F. Gao, “Swarm of micro flying robots in the wild,” Science Robotics , vol. 7, 2022

    2022

  19. [27]

    Racer: Rapid collaborative explo- ration with a decentralized multi-uav system,

    B. Zhou, H. Xu, and S. Shen, “Racer: Rapid collaborative explo- ration with a decentralized multi-uav system,” IEEE Transactions on Robotics, vol. 39, pp. 1816–1835, 2022

    2022

  20. [28]

    Adaptive tracking and perching for quadrotor in dynamic scenarios,

    Y . Gao, J. Ji, Q. Wang, R. Jin, Y . Lin, Z. Shang, Y . Cao, S. Shen, C. Xu, and F. Gao, “Adaptive tracking and perching for quadrotor in dynamic scenarios,” IEEE Transactions on Robotics , vol. 40, pp. 499– 519, 2023

    2023

  21. [29]

    Auto filmer: Autonomous aerial videography under human interaction,

    Z. Zhang, Y . Zhong, J. Guo, Q. Wang, C. Xu, and F. Gao, “Auto filmer: Autonomous aerial videography under human interaction,” IEEE Robotics and Automation Letters , vol. 8, pp. 784–791, 2023

    2023

  22. [30]

    Human motion diffusion model,

    G. Tevet, S. Raab, B. Gordon, Y . Shafir, D. Cohen-Or, and A. H. Bermano, “Human motion diffusion model,” ArXiv, vol. abs/2209.14916, 2022

    2022 arXiv

  23. [31]

    On the continuity of ro- tation representations in neural networks,

    Y . Zhou, C. Barnes, J. Lu, J. Yang, and H. Li, “On the continuity of ro- tation representations in neural networks,”2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pp. 5738–5746, 2018

    2019

  24. [32]

    Fast iterative region inflation for computing large 2-d/3-d convex regions of obstacle-free space,

    Q. Wang, Z. Wang, C. Xu, and F. Gao, “Fast iterative region inflation for computing large 2-d/3-d convex regions of obstacle-free space,” ArXiv, vol. abs/2403.02977, 2024

    2024 arXiv

  25. [33]

    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, pp. 3259–3278, 2021

    2021

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.