pith. sign in

arxiv: 1906.11669 · v1 · pith:COVKJWKTnew · submitted 2019-06-27 · 💻 cs.HC · cs.RO

Airways: Optimization-Based Planning of Quadrotor Trajectories according to High-Level User Goals

Christoph Gebhardt , Benjamin Hepp , Tobias Naegeli , Stefan Stevsic , Otmar Hilliges This is my paper

Pith reviewed 2026-05-25 14:44 UTC · model grok-4.3

classification 💻 cs.HC cs.RO
keywords quadrotor trajectoriesoptimization-based planninghigh-level user goalsaerial videographydrone racing3D design interfacecomputational design tool
0
0 comments X

The pith

An optimization method generates feasible quadrotor trajectories from high-level user goals without requiring control expertise.

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

The paper presents a design tool that lets non-expert users create quadrotor flight paths by specifying high-level objectives. An optimization procedure turns those objectives into concrete, flyable trajectories that respect real-world dynamics. A simple 3D interface supports quick editing and exploration of possible solutions. The same framework produces usable results across aerial videography, light painting, and racing scenarios. This setup removes the need for users to master low-level quadrotor constraints.

Core claim

The paper claims that an optimization-based planner can directly encode high-level human objectives into quadrotor trajectory generation, yielding paths that remain feasible under real dynamics and can be executed without the user specifying low-level control details or domain constraints.

What carries the argument

The optimization procedure that incorporates user objectives as terms in the planning problem to produce executable trajectories.

If this is right

  • Novice users can produce custom aerial videography paths.
  • The same interface supports robotic light-painting sequences.
  • Drone racing trajectories can be planned from high-level specifications.
  • Trajectories remain directly transferable to physical quadrotors.

Where Pith is reading between the lines

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

  • The approach could reduce the expertise barrier for operating other types of aerial robots.
  • Integration with live sensor feedback might allow on-the-fly trajectory adjustments.
  • Similar optimization encodings could apply to ground robots or multi-drone teams.

Load-bearing premise

The optimization step can reliably produce trajectories that stay feasible in real flight while exactly matching the user's stated high-level goals.

What would settle it

Execution of an output trajectory that either violates physical flight limits or visibly fails to meet the original high-level user objective.

Figures

Figures reproduced from arXiv: 1906.11669 by Benjamin Hepp, Christoph Gebhardt, Otmar Hilliges, Stefan Stevsic, Tobias Naegeli.

Figure 1
Figure 1. Figure 1: Interactive computational design of quadrotor trajectories: (A) user interface to specifiy keyframes and dynamics of [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: System workflow schematically. (1) User sketches keyframes. (2) An optimization method generates a feasible trajectory. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Planning of aerial video shots. (A) User specifies [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Our approximated quadrotor model with position [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figure 6
Figure 6. Figure 6: (A) camera direction pl and distance pd. (B) effect of minimizing camera angle error αerr w.r.t. the target rt in the center of the FOV of the camera. that are feasible and that are optimal in the sense of Eq. (5). While still relatively basic in functionality this already enables a variety of use-cases such as aerial light-shows and racing￾games as illustrated in the next section. Optimizing for Human Obj… view at source ↗
Figure 5
Figure 5. Figure 5: Same trajectory, optimized to only follow the [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: (A) Illustration of skewness error, where [PITH_FULL_IMAGE:figures/full_fig_p006_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Aerial camera shot of a toy castle. Top row: planned trajectory in our design tool. Bottom row: flown trajectory. [PITH_FULL_IMAGE:figures/full_fig_p007_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: (A) Handwritten input. (B) Initial feasible trajec [PITH_FULL_IMAGE:figures/full_fig_p008_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Two player aerial racing. User input is weighted [PITH_FULL_IMAGE:figures/full_fig_p008_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Multi target shot. Top row: frames of the video sequence shot by the onboard camera. Bottom row: according quadrotor [PITH_FULL_IMAGE:figures/full_fig_p009_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: A quadrotor in 3D with its flat outputs (position [PITH_FULL_IMAGE:figures/full_fig_p012_12.png] view at source ↗
read the original abstract

In this paper we propose a computational design tool that al-lows end-users to create advanced quadrotor trajectories witha variety of application scenarios in mind. Our algorithm al-lows novice users to create quadrotor based use-cases withoutrequiring deep knowledge in either quadrotor control or theunderlying constraints of the target domain. To achieve thisgoal we propose an optimization-based method that gener-ates feasible trajectories which can be flown in the real world.Furthermore, the method incorporates high-level human ob-jectives into the planning of flight trajectories. An easy touse 3D design tool allows for quick specification and edit-ing of trajectories as well as for intuitive exploration of theresulting solution space. We demonstrate the utility of our ap-proach in several real-world application scenarios, includingaerial-videography, robotic light-painting and drone racing.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 2 minor

Summary. The paper presents Airways, an optimization-based computational design tool for generating quadrotor trajectories from high-level user goals specified via an interactive 3D interface. The method produces trajectories claimed to be feasible for real-world flight while incorporating user objectives, demonstrated through applications in aerial videography, robotic light-painting, and drone racing. The central claim is that this allows novice users to create such trajectories without requiring deep expertise in quadrotor control or domain-specific constraints.

Significance. If the optimization reliably produces dynamically feasible trajectories that match user-specified high-level goals, the work could meaningfully advance HCI in robotics by reducing the expertise barrier for creative and applied quadrotor use cases. The interactive 3D tool and real-world demonstrations provide a concrete contribution to trajectory planning interfaces.

major comments (2)
  1. [Results / Evaluation] The abstract and introduction assert that generated trajectories remain feasible under real-world dynamics while exactly encoding high-level objectives, but the manuscript provides no quantitative validation metrics (e.g., tracking error, constraint violation rates, or success rates across trials) in the results or evaluation sections to support this for the demonstrated scenarios.
  2. [Method / Optimization] The optimization formulation is described at a high level but lacks explicit details on how user-specified objectives are encoded as cost terms or constraints (e.g., no equations showing the objective function or weighting of high-level goals versus feasibility terms), making it difficult to assess whether the method avoids reducing to prior self-cited approaches by construction.
minor comments (2)
  1. [Abstract] Abstract contains line-break artifacts such as 'al-lows' and inconsistent hyphenation ('aerial-videography' vs. 'aerial videography').
  2. [Introduction] The paper would benefit from clearer notation distinguishing between user-specified high-level goals and the low-level optimization variables.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment below and indicate planned revisions to the manuscript.

read point-by-point responses

  1. Referee: [Results / Evaluation] The abstract and introduction assert that generated trajectories remain feasible under real-world dynamics while exactly encoding high-level objectives, but the manuscript provides no quantitative validation metrics (e.g., tracking error, constraint violation rates, or success rates across trials) in the results or evaluation sections to support this for the demonstrated scenarios.

    Authors: The current manuscript emphasizes qualitative demonstrations of real-world flights in the application scenarios. We agree that the absence of quantitative metrics such as tracking error or success rates limits the strength of the feasibility claims. We will revise the evaluation section to incorporate quantitative metrics from the existing flight tests where available. revision: yes

  2. Referee: [Method / Optimization] The optimization formulation is described at a high level but lacks explicit details on how user-specified objectives are encoded as cost terms or constraints (e.g., no equations showing the objective function or weighting of high-level goals versus feasibility terms), making it difficult to assess whether the method avoids reducing to prior self-cited approaches by construction.

    Authors: The optimization is currently summarized at a high level. We will expand the method section in the revision to include the explicit objective function, cost terms encoding user goals, constraint formulations, and weighting details. This will also clarify distinctions from prior approaches. revision: yes

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper describes an optimization-based method for generating feasible quadrotor trajectories from high-level user objectives specified via an interactive 3D tool. No equations, parameter fits, or self-citations are shown that would reduce any claimed prediction or result to its own inputs by construction. The central claims rest on the formulation of an optimization procedure that encodes user goals while respecting real-world constraints, with validation through demonstrations in aerial videography, light-painting, and racing scenarios. This structure is self-contained against external benchmarks and does not invoke uniqueness theorems, ansatzes smuggled via citation, or renaming of known results as novel derivations.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

Abstract-only view yields limited visibility into parameters and assumptions; the method implicitly relies on standard trajectory optimization assumptions without explicit enumeration.

free parameters (1)
  • objective weighting factors
    Likely used to balance competing high-level goals such as smoothness versus speed; not quantified in abstract.
axioms (2)
  • domain assumption High-level user objectives can be encoded as cost terms in an optimization problem without loss of intent.
    Central to the claim that novice users need no deep domain knowledge.
  • domain assumption The resulting trajectories satisfy quadrotor dynamics and actuator limits sufficiently to be flown in the real world.
    Required for the feasibility guarantee stated in the abstract.

pith-pipeline@v0.9.0 · 5692 in / 1286 out tokens · 28558 ms · 2026-05-25T14:44:55.932449+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

40 extracted references · 40 canonical work pages

  1. [1]

    Javier Alonso-Mora, Tobias Naegeli, Roland Siegwart, and Paul Beardsley. 2015. Collision avoidance for aerial vehicles in multi-agent scenarios. Autonomous Robots 39, 1 (2015), 101–121. DOI: http://dx.doi.org/10.1007/s10514-015-9429-0

    work page doi:10.1007/s10514-015-9429-0 2015

  2. [2]

    ArsElectronica. 2012. Spaxels. (2012). http://www.aec.at/spaxels/

    work page 2012

  3. [3]

    Cannes Festival. 2012. New Directors’ Showcase. Video. (2012). https://youtu.be/cseTX_rW3uM 9

    work page 2012

  4. [4]

    Marc Christie, Rumesh Machap, Jean-Marie Normand, Patrick Olivier, and Jonathan Pickering. 2005. Virtual camera planning: A survey. In Smart Graphics. Springer, 40–52

    work page 2005

  5. [5]

    T.J. Diaz. 2015. Lights, drone... action. Spectrum, IEEE 52, 7 (July 2015), 36–41. DOI: http://dx.doi.org/10.1109/MSPEC.2015.7131693

    work page doi:10.1109/mspec.2015.7131693 2015

  6. [6]

    Nadeem Faiz, Sunil K Agrawal, and Richard M Murray

    work page

  7. [7]

    Journal of Guidance, Control, and Dynamics 24, 2 (2001), 219–227

    Trajectory planning of differentially flat systems with dynamics and inequalities. Journal of Guidance, Control, and Dynamics 24, 2 (2001), 219–227

    work page 2001

  8. [8]

    Tamar Flash and Neville Hogan. 1985. The coordination of arm movements: an experimentally confirmed mathematical model. The journal of Neuroscience 5, 7 (1985), 1688–1703

    work page 1985

  9. [9]

    Gleicher and Feng Liu

    Michael L. Gleicher and Feng Liu. 2008. Re-cinematography: Improving the Camerawork of Casual Video. ACM Trans. Multimedia Comput. Commun. Appl. 5, 1, Article 2 (Oct. 2008), 28 pages. DOI: http://dx.doi.org/10.1145/1404880.1404882

    work page doi:10.1145/1404880.1404882 2008

  10. [10]

    Grundmann, V

    M. Grundmann, V . Kwatra, and I. Essa. 2011. Auto-directed video stabilization with robust L1 optimal camera paths. In Computer Vision and Pattern Recognition (CVPR), 2011 IEEE Conference on. 225–232. DOI: http://dx.doi.org/10.1109/CVPR.2011.5995525

    work page doi:10.1109/cvpr.2011.5995525 2011

  11. [11]

    Keita Higuchi and Jun Rekimoto. 2012. Flying Head: Head-synchronized Unmanned Aerial Vehicle Control for Flying Telepresence. In SIGGRAPH Asia 2012 E-Tech (SA ’12). ACM, New York, NY , USA, 12:1—-12:2.DOI: http://dx.doi.org/10.1145/2407707.2407719

    work page doi:10.1145/2407707.2407719 2012

  12. [12]

    Keita Higuchi, Tetsuro Shimada, and Jun Rekimoto

    work page

  13. [13]

    In Augmented Human International Conference (AH ’11)

    Flying Sports Assistant: External Visual Imagery Representation for Sports Training. In Augmented Human International Conference (AH ’11). ACM, 7:1—-7:4. DOI:http://dx.doi.org/10.1145/1959826.1959833

    work page doi:10.1145/1959826.1959833

  14. [14]

    Niels Joubert, Mike Roberts, Anh Truong, Floraine Berthouzoz, and Pat Hanrahan. 2015. An Interactive Tool for Designing Quadrotor Camera Shots. ACM Trans. Graph. 34, 6, Article 238 (Oct. 2015), Article 238, 11 pages. DOI: http://dx.doi.org/10.1145/2816795.2818106

    work page doi:10.1145/2816795.2818106 2015

  15. [15]

    Jun Kato, Daisuke Sakamoto, and Takeo Igarashi. 2012. Phybots: A Toolkit for Making Robotic Things. In Proceedings of the Designing Interactive Systems Conference (DIS ’12). ACM, New York, NY , USA, 248–257. DOI: http://dx.doi.org/10.1145/2317956.2317996

    work page doi:10.1145/2317956.2317996 2012

  16. [16]

    Cohen, and Richard Szeliski

    Johannes Kopf, Michael F. Cohen, and Richard Szeliski

    work page

  17. [17]

    ACM Trans

    First-person Hyper-lapse Videos. ACM Trans. Graph. 33, 4, Article 78 (July 2014), 10 pages. DOI: http://dx.doi.org/10.1145/2601097.2601195

    work page doi:10.1145/2601097.2601195 2014

  18. [18]

    Taeyoung Lee, Melvin Leok, and N Harris McClamroch

    work page

  19. [19]

    Nonlinear robust tracking control of a quadrotor UA V on SE (3).Asian Journal of Control 15, 2 (2013), 391–408

    work page 2013

  20. [20]

    Tsai-Yen Li and Chung-Chiang Cheng. 2008. Real-Time Camera Planning for Navigation in Virtual Environments. In Smart Graphics, Andreas Butz, Brian Fisher, Antonio Krger, Patrick Olivier, and Marc Christie (Eds.). Lecture Notes in Computer Science, V ol. 5166. Springer Berlin Heidelberg, 118–129. DOI: http://dx.doi.org/10.1007/978-3-540-85412-8_11

    work page doi:10.1007/978-3-540-85412-8_11 2008

  21. [21]

    Kexi Liu, Daisuke Sakamoto, Masahiko Inami, and Takeo Igarashi. 2011. Roboshop: Multi-layered Sketching Interface for Robot Housework Assignment and Management. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’11). ACM, New York, NY , USA, 647–656.DOI: http://dx.doi.org/10.1145/1978942.1979035

    work page doi:10.1145/1978942.1979035 2011

  22. [22]

    Shuaicheng Liu, Lu Yuan, Ping Tan, and Jian Sun. 2013. Bundled Camera Paths for Video Stabilization. ACM Trans. Graph. 32, 4, Article 78 (July 2013), 10 pages. DOI:http://dx.doi.org/10.1145/2461912.2461995

    work page doi:10.1145/2461912.2461995 2013

  23. [23]

    Sergei Lupashin and Raffaello DAndrea. 2012. Adaptive fast open-loop maneuvers for quadrocopters. Autonomous Robots 33, 1-2 (April 2012), 89–102. http: //link.springer.com/10.1007/s10514-012-9289-9

    work page doi:10.1007/s10514-012-9289-9 2012

  24. [24]

    Robert Mahony, Vijay Kumar, and Peter Corke. 2012. Multirotor Aerial Vehicles: Modeling, Estimation, and Control of Quadrotor. IEEE Robotics & Automation Magazine 19, 3 (Sept. 2012), 20–32. http://ieeexplore.ieee.org/lpdocs/epic03/ wrapper.htm?arnumber=6289431

    work page 2012

  25. [25]

    Tobias Martin, Nobuyuki Umetani, and Bernd Bickel

    work page

  26. [26]

    ACM Trans

    OmniAD: Data-driven Omni-directional Aerodynamics. ACM Trans. Graph. 34, 4, Article 113 (July 2015), 12 pages. DOI: http://dx.doi.org/10.1145/2766919

    work page doi:10.1145/2766919 2015

  27. [27]

    Joseph V Mascelli. 1998. The five C’s of cinematography: motion picture filming techniques. Silman-James Press

    work page 1998

  28. [28]

    Lorenz Meier, Petri Tanskanen, Lionel Heng, Gim Hee Lee, Friedrich Fraundorfer, and Marc Pollefeys. 2012. PIXHAWK: A micro aerial vehicle design for autonomous flight using onboard computer vision. Autonomous Robots 33, 1-2 (Feb. 2012), 21–39. http: //link.springer.com/10.1007/s10514-012-9281-4

    work page doi:10.1007/s10514-012-9281-4 2012

  29. [29]

    Daniel Mellinger and Vijay Kumar. 2011. Minimum snap trajectory generation and control for quadrotors. In Robotics and Automation (ICRA), 2011 IEEE International Conference on. IEEE, 2520–2525

    work page 2011

  30. [30]

    Florian ’Floyd’ Mueller and Matthew Muirhead. 2015. Jogging with a Quadcopter. In Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems (CHI ’15). ACM, New York, NY , USA, 2023–2032. DOI: http://dx.doi.org/10.1145/2702123.2702472 10

    work page doi:10.1145/2702123.2702472 2015

  31. [31]

    Tobias N ¨ageli, Christian Conte, Alexander Domahidi, Manfred Morari, and Otmar Hilliges. 2014. Environment-independent formation flight for micro aerial vehicles. In Intelligent Robots and Systems (IROS 2014), 2014 IEEE/RSJ International Conference on. 1141–1146. DOI: http://dx.doi.org/10.1109/IROS.2014.6942701

    work page doi:10.1109/iros.2014.6942701 2014

  32. [32]

    Tayyab Naseer, J Sturm, and D Cremers. 2013. Followme: Person following and gesture recognition with a quadrocopter. Proc. IROS (2013). https://vision.in. tum.de/_media/spezial/bib/naseer2013iros.pdf

    work page 2013

  33. [33]

    Kei Nitta, Keita Higuchi, and Jun Rekimoto. 2014. HoverBall: Augmented Sports with a Flying Ball. In Augmented Human International Conference (AH ’14) (AH ’14). ACM, New York, NY , USA, 13:1—-13:4.DOI: http://dx.doi.org/10.1145/2582051.2582064

    work page doi:10.1145/2582051.2582064 2014

  34. [34]

    Daisuke Sakamoto, Koichiro Honda, Masahiko Inami, and Takeo Igarashi. 2009. Sketch and run. In ACM SIGCHI. ACM Press, New York, New York, USA, 197. http: //dl.acm.org/citation.cfm?id=1518701.1518733

    work page arXiv 2009

  35. [35]

    J ¨urgen Scheible, Achim Hoth, Julian Saal, and Haifeng Su. 2013. Displaydrone: a flying robot based interactive display. In ACM International Symposium on Pervasive Displays (PerDis ’13). ACM Press, New York, New York, USA, 49. http: //dl.acm.org/citation.cfm?id=2491568.2491580

    work page arXiv 2013

  36. [36]

    Yuta Sugiura, Diasuke Sakamoto, Anusha Withana, Masahiko Inami, and Takeo Igarashi. 2010. Cooking with robots. In ACM SIGCHI. ACM Press, New York, New York, USA, 2427. http: //dl.acm.org/citation.cfm?id=1753326.1753693

    work page arXiv 2010

  37. [37]

    Nobuyuki Umetani, Yuki Koyama, Ryan Schmidt, and Takeo Igarashi. 2014. Pteromys: Interactive Design and Optimization of Free-formed Free-flight Model Airplanes. ACM Trans. Graph. 33, 4, Article 65 (July 2014), 10 pages. DOI: http://dx.doi.org/10.1145/2601097.2601129

    work page doi:10.1145/2601097.2601129 2014

  38. [38]

    I-Cheng Yeh, Chao-Hung Lin, Hung-Jen Chien, and Tong-Yee Lee. 2011. Efficient camera path planning algorithm for human motion overview. Computer Animation and Virtual Worlds22, 2-3 (2011), 239–250. DOI:http://dx.doi.org/10.1002/cav.398

    work page doi:10.1002/cav.398 2011

  39. [39]

    Shigeo Yoshida, Takumi Shirokura, Yuta Sugiura, Daisuke Sakamoto, Tetsuo Ono, Masahiko Inami, and Takeo Igarashi. 2015. RoboJockey: Designing an Entertainment Experience with Robots. IEEE computer graphics and applications 1 (Jan. 2015), 1. http://www.computer.org/csdl/mags/cg/preprint/ 07005367-abs.html

    work page 2015

  40. [40]

    Shengdong Zhao, Koichi Nakamura, Kentaro Ishii, and Takeo Igarashi. 2009. Magic Cards: A Paper Tag Interface for Implicit Robot Control. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’09). ACM, New York, NY , USA, 173–182. DOI:http://dx.doi.org/10.1145/1518701.1518730 APPENDIX In the work proposed here we use an approx...

    work page doi:10.1145/1518701.1518730 2009