pith. sign in

arxiv: 1907.00520 · v1 · pith:I4KW3KL7new · submitted 2019-07-01 · 💻 cs.RO

Teach-Repeat-Replan: A Complete and Robust System for Aggressive Flight in Complex Environments

Fei Gao , Luqi Wang , Boyu Zhou , Luxin Han , Jie Pan , Shaojie Shen This is my paper

Pith reviewed 2026-05-25 12:23 UTC · model grok-4.3

classification 💻 cs.RO
keywords teach-repeat-replanquadrotor motion planningaggressive flighttrajectory smoothinglocal replanningautonomous navigationkinodynamic feasibilityteach and repeat
0
0 comments X

The pith

The teach-repeat-replan system converts any jerky human teaching path into a smooth, safe, topologically equivalent trajectory and adds local replanning to avoid unmapped or moving obstacles.

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

The paper presents a complete motion planning system for autonomous quadrotors that starts from a human-piloted teaching path and converts it into a smooth repeating trajectory while preserving the topological structure of the user's intended mission. The conversion produces a trajectory that is guaranteed smooth, safe, and kinodynamically feasible with human-preferable aggressiveness. A sliding-window local perception module then generates onboard re-plans to handle unmapped or dynamic obstacles that appear after the initial teaching. The full pipeline is integrated with global and local perception plus localization modules to support aggressive flights in complex indoor and outdoor environments, and the components are released as open-source ROS packages.

Core claim

By converting an arbitrarily jerky human teaching trajectory into a topologically equivalent smooth, safe, and kinodynamically feasible repeating trajectory, and by adding sliding-windowed local perception and replanning, the teach-repeat-replan system captures user intention, enables aggressive quadrotor flight, and safely avoids unmapped or moving obstacles.

What carries the argument

The teach-repeat-replan pipeline that converts a human teaching path to a topologically equivalent smooth trajectory and performs sliding-window local replanning.

If this is right

  • The system supports aggressive flights for infrastructure inspection, aerial transport, and search-and-rescue without requiring the drone to follow the teaching path exactly.
  • Local replanning allows continued operation when environments change after the teaching phase.
  • The open-source release provides reusable modules for perception, localization, and planning on other quadrotor platforms.
  • Demonstrations confirm the approach works across both indoor and outdoor challenging settings.

Where Pith is reading between the lines

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

  • The same conversion-plus-local-replan structure could apply to other robots where preserving topological mission intent matters more than exact path following.
  • Reducing dependence on complete global maps through local replanning may help scale operations to larger or less mapped areas.
  • Tighter coupling between the global conversion step and the local planner could improve performance when obstacles move quickly.

Load-bearing premise

Every human teaching path always admits a topologically equivalent smooth safe trajectory that can be generated while preserving the intended mission structure, and sliding-window perception is always enough to produce collision-free local replans.

What would settle it

A flight test in which the conversion step produces no kinodynamically feasible trajectory or the local replanner fails to prevent collision with an unmapped or moving obstacle.

Figures

Figures reproduced from arXiv: 1907.00520 by Boyu Zhou, Fei Gao, Jie Pan, Luqi Wang, Luxin Han, Shaojie Shen.

Figure 1
Figure 1. Figure 1: Experiments in a challenging indoor drone racing site and an outdoor [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The software architecture of our quadrotor system. Global mapping, planning, and visualization are running on a ground station, while state estimation, [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The hardware setting of our autonomous drone system. [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: An illustration of free space captured by an axis-aligned cube and [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The comparison of an axis-aligned cube and a general convex [PITH_FULL_IMAGE:figures/full_fig_p004_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: An illustration of the convex cluster inflation. In (a) and (b), all qualified neighbor voxels are added to the convex cluster. In (c) and (d), since an occupied voxel occludes a ray (the green arrow) to one of the clustered voxels, the testing voxel (in yellow) is excluded to the convex cluster. C ∗ contains newly added voxels which preserve the convexity. The cluster inflation starts by adding the seed v… view at source ↗
Figure 7
Figure 7. Figure 7: The flight corridor generation process. Red dots are coordinates along the teaching trajectory. (b), a new convex polyhedron is generated and added [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: The effect of the temporal optimization. [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: An illustration of colliding with obstacles when there are significant [PITH_FULL_IMAGE:figures/full_fig_p011_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: The local occupancy map its corresponding ESDF map visualized [PITH_FULL_IMAGE:figures/full_fig_p011_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: An illustration of the online re-planning mechanism. The blue and green curves are the global trajectory and the actual flight path of the drone, [PITH_FULL_IMAGE:figures/full_fig_p012_11.png] view at source ↗
Figure 13
Figure 13. Figure 13: The flight corridor generated with and without the initialization [PITH_FULL_IMAGE:figures/full_fig_p013_13.png] view at source ↗
Figure 12
Figure 12. Figure 12: The trajectory generated in a complex simulated environment. The [PITH_FULL_IMAGE:figures/full_fig_p013_12.png] view at source ↗
Figure 14
Figure 14. Figure 14: The comparison of trajectories optimized by different methods. [PITH_FULL_IMAGE:figures/full_fig_p014_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: The experimental set-up of the fast indoor drone racing flights. (a), [PITH_FULL_IMAGE:figures/full_fig_p015_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: Snapshots of the fast autonomous flight in a static indoor environ [PITH_FULL_IMAGE:figures/full_fig_p015_16.png] view at source ↗
Figure 21
Figure 21. Figure 21: The repeating trajectory in outdoor experiments, trial 1. Marks are [PITH_FULL_IMAGE:figures/full_fig_p016_21.png] view at source ↗
Figure 19
Figure 19. Figure 19: Indoor flight in a dynamic environment. In (a) and (b), the unmapped [PITH_FULL_IMAGE:figures/full_fig_p016_19.png] view at source ↗
Figure 20
Figure 20. Figure 20: Snapshots of the experiments in outdoor environments. [PITH_FULL_IMAGE:figures/full_fig_p016_20.png] view at source ↗
Figure 23
Figure 23. Figure 23: Profiles of the desired and estimated position and velocity. The [PITH_FULL_IMAGE:figures/full_fig_p017_23.png] view at source ↗
read the original abstract

In this paper, we propose a complete and robust motion planning system for the aggressive flight of autonomous quadrotors. The proposed method is built upon on a classical teach-and-repeat framework, which is widely adopted in infrastructure inspection, aerial transportation, and search-and-rescue. For these applications, human's intention is essential to decide the topological structure of the flight trajectory of the drone. However, poor teaching trajectories and changing environments prevent a simple teach-and-repeat system from being applied flexibly and robustly. In this paper, instead of commanding the drone to precisely follow a teaching trajectory, we propose a method to automatically convert a human-piloted trajectory, which can be arbitrarily jerky, to a topologically equivalent one. The generated trajectory is guaranteed to be smooth, safe, and kinodynamically feasible, with a human preferable aggressiveness. Also, to avoid unmapped or dynamic obstacles during flights, a sliding-windowed local perception and re-planning method are introduced to our system, to generate safe local trajectories onboard. We name our system as teach-repeat-replan. It can capture users' intention of a flight mission, convert an arbitrarily jerky teaching path to a smooth repeating trajectory, and generate safe local re-plans to avoid unmapped or moving obstacles. The proposed planning system is integrated into a complete autonomous quadrotor with global and local perception and localization sub-modules. Our system is validated by performing aggressive flights in challenging indoor/outdoor environments. We release all components in our quadrotor system as open-source ros-packages.

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 / 1 minor

Summary. The paper proposes the teach-repeat-replan system for aggressive autonomous quadrotor flight. It converts arbitrarily jerky human teaching trajectories into smooth, safe, and kinodynamically feasible repeating trajectories that preserve the user's intended topological structure. A sliding-window local perception and replanning method handles unmapped or dynamic obstacles. The system is integrated with perception and localization modules and validated through aggressive flights in challenging indoor and outdoor environments, with all components released as open-source ROS packages.

Significance. If the central claims hold, this work offers a practical and robust approach for applications requiring human-guided topological structure in flight missions, such as infrastructure inspection and search-and-rescue, by addressing limitations of simple teach-and-repeat in changing environments. The open-source release of the full quadrotor system is a notable strength that facilitates reproducibility and further research.

major comments (2)
  1. [Abstract] Abstract: The assertion that any 'arbitrarily jerky' teaching path can be converted to a 'topologically equivalent' trajectory that is 'guaranteed to be smooth, safe, and kinodynamically feasible' lacks a formal derivation, completeness argument, or explicit conditions on obstacle density, path clearance, and bounded sensing range. This is load-bearing for the robustness claim and directly engages the risk that B-spline/polynomial smoothing or sliding-window replanning may exit the original homotopy class in narrow or topologically complex environments.
  2. [Teach-repeat module] Teach-repeat module description: No quantitative metrics (success rate of topological preservation, trajectory deviation, or failure cases under reduced clearance) or comparison against baselines are supplied to support the conversion operator, leaving the experimental validation of 'aggressive flights in challenging environments' without measurable evidence.
minor comments (1)
  1. [Abstract] Abstract: The final sentence introducing the system name could be merged with the preceding claim for improved flow.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment point by point below, with revisions proposed where they strengthen the manuscript without misrepresenting our contributions.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The assertion that any 'arbitrarily jerky' teaching path can be converted to a 'topologically equivalent' trajectory that is 'guaranteed to be smooth, safe, and kinodynamically feasible' lacks a formal derivation, completeness argument, or explicit conditions on obstacle density, path clearance, and bounded sensing range. This is load-bearing for the robustness claim and directly engages the risk that B-spline/polynomial smoothing or sliding-window replanning may exit the original homotopy class in narrow or topologically complex environments.

    Authors: The teach-repeat module constructs a safe flight corridor from the teaching trajectory and optimizes a B-spline within it, which by construction remains in the same homotopy class when the teaching path maintains clearance larger than the quadrotor radius plus a safety margin. The abstract statement is therefore conditioned on this clearance assumption and bounded sensing range for local replanning. We agree a more explicit statement of these conditions and a short paragraph on why the corridor method precludes homotopy-class exit would improve clarity. We will revise the abstract and add this discussion to the teach-repeat section. revision: yes

  2. Referee: [Teach-repeat module] Teach-repeat module description: No quantitative metrics (success rate of topological preservation, trajectory deviation, or failure cases under reduced clearance) or comparison against baselines are supplied to support the conversion operator, leaving the experimental validation of 'aggressive flights in challenging environments' without measurable evidence.

    Authors: The current experiments demonstrate end-to-end system success in aggressive real-world flights, which implicitly tests the teach-repeat conversion. We nevertheless agree that dedicated quantitative evaluation of the conversion operator itself would strengthen the claims. In revision we will add metrics including mean/max trajectory deviation from the teaching path, success rate of topological preservation over repeated trials, and direct comparison against baseline B-spline smoothing without corridor constraints, including failure cases at reduced clearance. revision: yes

Circularity Check

0 steps flagged

No circularity; system architecture claims rest on algorithmic modules, not self-referential definitions or fits.

full rationale

The paper presents a teach-repeat-replan architecture for quadrotor trajectory conversion and local replanning. No equations, fitted parameters, or predictions appear in the provided text. The central claim of producing a topologically equivalent smooth trajectory is presented as a property of the proposed conversion method rather than derived from or equivalent to its inputs by construction. Self-citations are not load-bearing for any uniqueness theorem. The open-source release allows external verification, confirming the derivation chain is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review; no mathematical derivations, fitted constants, or new entities are visible. The system relies on standard robotics assumptions (kinodynamic feasibility, topological equivalence preservation) that are not enumerated as free parameters or axioms here.

pith-pipeline@v0.9.0 · 5822 in / 1193 out tokens · 45948 ms · 2026-05-25T12:23:31.305385+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

46 extracted references · 46 canonical work pages · 2 internal anchors

  1. [1]

    Optimal trajectory generation for quadrotor teach-and-repeat,

    F. Gao, L. Wang, K. Wang, W. Wu, B. Zhou, L. Han, and S. Shen, “Optimal trajectory generation for quadrotor teach-and-repeat,” IEEE Robotics and Automation Letters (RA-L) , 2019

    work page 2019

  2. [2]

    Visual-Inertial Teach and Repeat for Aerial Inspection

    M. Fehr, T. Schneider, M. Dymczyk, J. Sturm, and R. Siegwart, “Visual-inertial teach and repeat for aerial inspection,” arXiv preprint arXiv:1803.09650, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  3. [3]

    Visual teach and repeat for long-range rover autonomy,

    P. Furgale and T. D. Barfoot, “Visual teach and repeat for long-range rover autonomy,” Journal of Field Robotics, vol. 27, no. 5, pp. 534–560, 2010

    work page 2010

  4. [4]

    FIESTA: Fast Incremental Euclidean Distance Fields for Online Motion Planning of Aerial Robots

    L. Han, F. Gao, B. Zhou, and S. Shen, “Fiesta: Fast incremental euclidean distance fields for online motion planning of aerial robots,” arXiv preprint arXiv:1903.02144 , 2019

    work page internal anchor Pith review Pith/arXiv arXiv 1903

  5. [5]

    Robust and effi- cient quadrotor trajectory generation for fast autonomous flight,

    B. Zhou, F. Gao, L. Wang, C. Liu, and S. Shen, “Robust and effi- cient quadrotor trajectory generation for fast autonomous flight,” IEEE Robotics and Automation Letters (RA-L) , 2019

    work page 2019

  6. [6]

    Lidar-based teach-and-repeat of mobile robot trajectories,

    C. Sprunk, G. D. Tipaldi, A. Cherubini, and W. Burgard, “Lidar-based teach-and-repeat of mobile robot trajectories,” in Proc. of the IEEE/RSJ Intl. Conf. on Intell. Robots and Syst.(IROS) , 2013, pp. 3144–3149

    work page 2013

  7. [7]

    There and back again-dealing with highly-dynamic scenes and long-term change during topological/metric route following,

    P. Furgale, P. Kr ¨usi, F. Pomerleau, U. Schwesinger, F. Colas, and R. Siegwart, “There and back again-dealing with highly-dynamic scenes and long-term change during topological/metric route following,” in ICRA14 Workshop on Modelling, Estimation, Perception, and Control of All Terrain Mobile Robots , 2014

    work page 2014

  8. [8]

    Lighting-invariant adaptive route following using iterative closest point matching,

    P. Kr ¨usi, B. B ¨ucheler, F. Pomerleau, U. Schwesinger, R. Siegwart, and P. Furgale, “Lighting-invariant adaptive route following using iterative closest point matching,” Journal of Field Robotics , vol. 32, no. 4, pp. 534–564, 2015

    work page 2015

  9. [9]

    Visual teach and repeat, repeat: Iterative learning control to improve mobile robot path tracking in challenging outdoor environments,

    C. J. Ostafew, A. P. Schoellig, and T. D. Barfoot, “Visual teach and repeat, repeat: Iterative learning control to improve mobile robot path tracking in challenging outdoor environments,” in Proc. of the IEEE/RSJ Intl. Conf. on Intell. Robots and Syst.(IROS) , 2013, pp. 176–181

    work page 2013

  10. [10]

    Bridging the appearance gap: Multi-experience localization for long-term visual teach and repeat,

    M. Paton, K. MacTavish, M. Warren, and T. D. Barfoot, “Bridging the appearance gap: Multi-experience localization for long-term visual teach and repeat,” in Proc. of the IEEE/RSJ Intl. Conf. on Intell. Robots and Syst.(IROS), 2016, pp. 1918–1925

    work page 2016

  11. [11]

    It’s not easy seeing green: Lighting-resistant stereo visual teach & repeat using color-constant images,

    M. Paton, K. MacTavish, C. J. Ostafew, and T. D. Barfoot, “It’s not easy seeing green: Lighting-resistant stereo visual teach & repeat using color-constant images,” in Proc. of the IEEE Intl. Conf. on Robot. and Autom. (ICRA), 2015, pp. 1519–1526

    work page 2015

  12. [12]

    It’s like d ´ej`a vu all over again: Learning place-dependent terrain assessment for visual teach and repeat,

    L.-P. Berczi and T. D. Barfoot, “It’s like d ´ej`a vu all over again: Learning place-dependent terrain assessment for visual teach and repeat,” in Proc. of the IEEE/RSJ Intl. Conf. on Intell. Robots and Syst.(IROS) , 2016, pp. 3973–3980

    work page 2016

  13. [13]

    Minimum snap trajectory generation and control for quadrotors,

    D. Mellinger and V . Kumar, “Minimum snap trajectory generation and control for quadrotors,” in Proc. of the IEEE Intl. Conf. on Robot. and Autom. (ICRA), Shanghai, China, May 2011, pp. 2520–2525

    work page 2011

  14. [14]

    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 Proc. of the Intl. Sym. of Robot. Research (ISRR) , Dec. 2013, pp. 649–666

    work page 2013

  15. [15]

    Online safe trajectory generation for quadrotors using fast marching method and bernstein basis polyno- mial,

    F. Gao, W. Wu, Y . Lin, and S. Shen, “Online safe trajectory generation for quadrotors using fast marching method and bernstein basis polyno- mial,” in Proc. of the IEEE Intl. Conf. on Robot. and Autom. (ICRA) , Brisbane, Australia, May 2018

    work page 2018

  16. [16]

    Online quadrotor trajectory generation and autonomous navigation on point clouds,

    F. Gao and S. Shen, “Online quadrotor trajectory generation and autonomous navigation on point clouds,” in Proc. of the IEEE Intl. Sym. on Safety, Security, and Rescue Robotics (SSRR) , lausanne, switzerland, 2016, pp. 139–146

    work page 2016

  17. [17]

    Flying on point clouds: Online trajectory generation and autonomous navigation for quadrotors in cluttered environments,

    F. Gao, W. Wu, W. Gao, and S. Shen, “Flying on point clouds: Online trajectory generation and autonomous navigation for quadrotors in cluttered environments,” Journal of Field Robotics , 2018

    work page 2018

  18. [18]

    Planning dynamically feasible trajectories for quadrotors using safe flight corridors in 3-d complex environments,

    S. Liu, M. Watterson, K. Mohta, K. Sun, S. Bhattacharya, C. J. Taylor, and V . Kumar, “Planning dynamically feasible trajectories for quadrotors using safe flight corridors in 3-d complex environments,” IEEE Robotics and Automation Letters (RA-L) , pp. 1688–1695, 2017

    work page 2017

  19. [19]

    Computing large convex regions of obstacle- free space through semidefinite programming,

    R. Deits and R. Tedrake, “Computing large convex regions of obstacle- free space through semidefinite programming,” in Algorithmic Founda- tions of Robotics XI . Springer, 2015, vol. 107, pp. 109–124

    work page 2015

  20. [20]

    Chomp: Gradient optimization techniques for efficient motion planning,

    N. Ratliff, M. Zucker, J. A. Bagnell, and S. Srinivasa, “Chomp: Gradient optimization techniques for efficient motion planning,” in Proc. of the IEEE Intl. Conf. on Robot. and Autom. (ICRA) , May 2009, pp. 489–494

    work page 2009

  21. [21]

    Continuous-time trajectory optimization for online uav replan- ning,

    H. Oleynikova, M. Burri, Z. Taylor, J. Nieto, R. Siegwart, and E. Gal- ceran, “Continuous-time trajectory optimization for online uav replan- ning,” in Proc. of the IEEE/RSJ Intl. Conf. on Intell. Robots and Syst.(IROS), Daejeon, Korea, Oct. 2016, pp. 5332–5339. 18

    work page 2016

  22. [22]

    Autonomous aerial navigation using monocular visual-inertial fusion,

    Y . Lin, F. Gao, T. Qin, W. Gao, T. Liu, W. Wu, and S. Shen, “Autonomous aerial navigation using monocular visual-inertial fusion,” Journal of Field Robotics , vol. 35, no. 1, pp. 23–51, 2018

    work page 2018

  23. [23]

    An open-source system for vision-based micro-aerial vehicle mapping, planning, and flight in cluttered environ- ments,

    H. Oleynikova, C. Lanegger, Z. Taylor, M. Pantic, M. Alexander, R. Siegwart, and J. Nieto, “An open-source system for vision-based micro-aerial vehicle mapping, planning, and flight in cluttered environ- ments,” arXiv preprint arXiv:1812.03892 , 2019

    work page arXiv 2019

  24. [24]

    H. M. Choset, S. Hutchinson, K. M. Lynch, G. Kantor, W. Burgard, L. E. Kavraki, and S. Thrun, Principles of robot motion: theory, algorithms, and implementation. MIT press, 2005

    work page 2005

  25. [25]

    Generating dynamically feasible trajec- tories for quadrotor cameras,

    M. Roberts and P. Hanrahan, “Generating dynamically feasible trajec- tories for quadrotor cameras,” ACM Transactions on Graphics (TOG) , vol. 35, no. 4, p. 61, 2016

    work page 2016

  26. [26]

    Near minimum-time trajectories for quadrotor uavs in complex environments,

    J. Jamieson and J. Biggs, “Near minimum-time trajectories for quadrotor uavs in complex environments,” in Proc. of the IEEE/RSJ Intl. Conf. on Intell. Robots and Syst.(IROS) , 2016, pp. 1550–1555

    work page 2016

  27. [27]

    Time-optimal path tracking for robots: A convex optimization approach,

    D. Verscheure, B. Demeulenaere, J. Swevers, J. De Schutter, and M. Diehl, “Time-optimal path tracking for robots: A convex optimization approach,” IEEE Transactions on Automatic Control, vol. 54, no. 10, pp. 2318–2327, 2009

    work page 2009

  28. [28]

    A general, fast, and robust implementation of the time-optimal path parameterization algorithm,

    Q.-C. Pham, “A general, fast, and robust implementation of the time-optimal path parameterization algorithm,” IEEE Transactions on Robotics, vol. 30, no. 6, pp. 1533–1540, 2014

    work page 2014

  29. [29]

    A new approach to time-optimal path parameterization based on reachability analysis,

    H. Pham and Q.-C. Pham, “A new approach to time-optimal path parameterization based on reachability analysis,” IEEE Transactions on Robotics, vol. 34, no. 3, pp. 645–659, 2018

    work page 2018

  30. [30]

    Coordinate descent algorithms,

    S. J. Wright, “Coordinate descent algorithms,” Mathematical Program- ming, vol. 151, no. 1, pp. 3–34, 2015

    work page 2015

  31. [31]

    Geometric tracking control of a quadrotor uav on se (3),

    T. Lee, M. Leoky, and N. H. McClamroch, “Geometric tracking control of a quadrotor uav on se (3),” in Proc. of the IEEE Control and Decision Conf. (CDC), Atlanta, GA, Dec. 2010, pp. 5420–5425

    work page 2010

  32. [32]

    Vins-mono: A robust and versatile monocular visual-inertial state estimator,

    T. Qin, P. Li, and S. Shen, “Vins-mono: A robust and versatile monocular visual-inertial state estimator,” IEEE Transactions on Robotics , vol. 34, no. 4, pp. 1004–1020, 2018

    work page 2018

  33. [33]

    Real-time scalable dense surfel mapping,

    K. Wang, F. Gao, and S. Shen, “Real-time scalable dense surfel mapping,” in Proc. of the IEEE Intl. Conf. on Robot. and Autom. (ICRA), 2019

    work page 2019

  34. [34]

    Topomap: Topological mapping and navigation based on visual slam maps,

    F. Blochliger, M. Fehr, M. Dymczyk, T. Schneider, and R. Siegwart, “Topomap: Topological mapping and navigation based on visual slam maps,” in Proc. of the IEEE Intl. Conf. on Robot. and Autom. (ICRA) , 2018, pp. 1–9

    work page 2018

  35. [35]

    Lay, Convex sets and their applications

    S. Lay, Convex sets and their applications . Courier Corporation, 2007

    work page 2007

  36. [36]

    The quickhull algorithm for convex hulls,

    C. B. Barber, D. P. Dobkin, D. P. Dobkin, and H. Huhdanpaa, “The quickhull algorithm for convex hulls,” ACM Transactions on Mathemat- ical Software (TOMS) , vol. 22, no. 4, pp. 469–483, 1996

    work page 1996

  37. [37]

    Double description method revisited,

    K. Fukuda and A. Prodon, “Double description method revisited,” in Franco-Japanese and Franco-Chinese Conference on Combinatorics and Computer Science . Springer, 1995, pp. 91–111

    work page 1995

  38. [38]

    Optimal time allocation for quadrotor trajectory generation,

    F. Gao, W. Wu, J. Pan, B. Zhou, and S. Shen, “Optimal time allocation for quadrotor trajectory generation,” in Proc. of the IEEE/RSJ Intl. Conf. on Intell. Robots and Syst.(IROS) , Madrid, Spain, Oct 2018

    work page 2018

  39. [39]

    Distance transforms of sampled functions,

    P. F. Felzenszwalb and D. P. Huttenlocher, “Distance transforms of sampled functions,” Theory of computing , vol. 8, no. 1, pp. 415–428, 2012

    work page 2012

  40. [40]

    Incremental distance transforms (idt),

    T. Schouten and E. L. van den Broek, “Incremental distance transforms (idt),” in 2010 20th International Conference on Pattern Recognition . IEEE, 2010, pp. 237–240

    work page 2010

  41. [41]

    Chisel: Real time large scale 3d reconstruction onboard a mobile device using spatially hashed signed distance fields

    M. Klingensmith, I. Dryanovski, S. Srinivasa, and J. Xiao, “Chisel: Real time large scale 3d reconstruction onboard a mobile device using spatially hashed signed distance fields.” inProc. of Robot.: Sci. and Syst. (RSS), 2015

    work page 2015

  42. [42]

    Subroutine package for calculating with b-splines

    C. de Boor, “Subroutine package for calculating with b-splines.” Los Alamos Scientific Lab., N. Mex., Tech. Rep., 1971

    work page 1971

  43. [43]

    Elastic bands: Connecting path planning and control,

    S. Quinlan and O. Khatib, “Elastic bands: Connecting path planning and control,” 1993, pp. 802–807

    work page 1993

  44. [44]

    A convex optimization approach to smooth trajectories for motion planning with car-like robots,

    Z. Zhu, E. Schmerling, and M. Pavone, “A convex optimization approach to smooth trajectories for motion planning with car-like robots,” in Proc. of the IEEE Control and Decision Conf. (CDC) , 2015, pp. 835–842

    work page 2015

  45. [45]

    Gradient-based online safe trajectory generation for quadrotor flight in complex environments,

    F. Gao, Y . Lin, and S. Shen, “Gradient-based online safe trajectory generation for quadrotor flight in complex environments,” in Proc. of the IEEE/RSJ Intl. Conf. on Intell. Robots and Syst.(IROS) , Sept 2017, pp. 3681–3688

    work page 2017

  46. [46]

    Real- time trajectory replanning for mavs using uniform b-splines and a 3d circular buffer,

    V . Usenko, L. von Stumberg, A. Pangercic, and D. Cremers, “Real- time trajectory replanning for mavs using uniform b-splines and a 3d circular buffer,” in Proc. of the IEEE/RSJ Intl. Conf. on Intell. Robots and Syst.(IROS), 2017, pp. 215–222

    work page 2017