Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Swarm Intelligence in Collision-free Formation Control for Multi-UAV Systems with 3D Obstacle Avoidance Maneuvers

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read This paper proposes a semi-centralized controller that combines a probabilistic Lloyd's algorithm for formation with distributed potential fields for collision and obstacle avoidance, extending pigeon-inspired maneuvers to 3D for…

desk verdict The 3D avoidance term is rotation-invariant, so the central claim of a 3D maneuver collapses; the formation control part is standard and the simulations don't compensate. read the letter →

arxiv 2412.12437 v1 pith:HRSDJXUQ submitted 2024-12-17 cs.RO

Reza Ahmadvand , Sarah Sharif , Yaser Banad This is my paper · ORCID
classification cs.RO
keywords multi-UAVsystemsswarmintelligenceformationcontrolcollisionavoidanceobstaclecentroidalVoronoitessellationLloyd'salgorithmdistributed
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 aims to show that a multi-UAV swarm can keep a formation, avoid hitting each other, and steer around static and moving obstacles in 3D using a controller that mixes two inspirations: territorial tilapia fish for formation geometry and pigeons for obstacle maneuvers. The formation layer is semi-centralized, using a probabilistic Lloyd's algorithm to compute centroidal Voronoi positions that the drones track. The safety layer is distributed, with each drone applying local potential-field controls for inter-vehicle collision avoidance and obstacle avoidance. The proposed 3D obstacle term rotates the obstacle-relative displacement with yaw and pitch rotation matrices inside a potential function, which is meant to give drones extra maneuverability beyond planar avoidance. The paper reports simulations of 8 and 12 drones, including formation change and recovery between buildings and around static and dynamic obstacles, and claims these runs stay collision-free without consensus-based control.

What carries the argument

The carrying mechanisms are the probabilistic Lloyd's algorithm for centroidal Voronoi tessellation, which produces the formation targets, and the rotational obstacle potential $U_r=\frac{k_r}{2}\|T_{rx}T_{ry}T_{rz}(\boldsymbol{p}_i-\boldsymbol{o}_k)\|^2$ with yaw and pitch rotation matrices $T_{ry}$, $T_{rz}$ and an identity matrix $T_{rx}$; the distance-dependent angle $\alpha$ is supposed to shape the avoidance force as the drone approaches an obstacle. These are combined through the superposition control law $\boldsymbol{u}_i=\boldsymbol{u}_{fi}+\boldsymbol{u}_{ci}+\boldsymbol{u}_{oi}$, with the obstacle term gated by two detection conditions: range within $r_d+r_{ok}$ and the obstacle lying inside a conical field of view around the flight direction.

What would settle it

Evaluate the rotational potential in Eq. (20) for any detected obstacle: because $T_{rx},T_{ry},T_{rz}$ are rotations, $\|T_{rx}T_{ry}T_{rz}(\boldsymbol{p}_i-\boldsymbol{o}_k)\|=\|\boldsymbol{p}_i-\boldsymbol{o}_k\|$, so $U_r=\frac{k_r}{2}\|\boldsymbol{p}_i-\boldsymbol{o}_k\|^2$ and its gradient is the radial force $k_r(\boldsymbol{p}_i-\boldsymbol{o}_k)$, independent of the angle $\alpha$. A simulation with only the obstacle term activated should therefore show no lateral or vertical steering around an off-axis obstacle, which would contradict the reported 3D obstacle avoidance maneuver.

Watch

Extended reading notes

Core claim

The central claim is that formation planning and safety can be cleanly separated in a multi-UAV swarm: Lloyd's algorithm on a Voronoi cost function supplies each drone with an optimal target position, while a superposition of three control terms—formation tracking, inter-vehicle repulsion, and an obstacle potential—keeps the flight collision-free. The distinct novelty is the obstacle term, written as a rotational potential $U_r=\frac{k_r}{2}\|T_{rx}T_{ry}T_{rz}(\boldsymbol{p}_i-\boldsymbol{o}_k)\|^2$, whose gradient is intended to push drones around obstacles in 3D rather than simply away from them; the rotation angle $\alpha$ decreases from $\pi/2$ at the safety radius to zero at the detection radius. Applied to 8 and then 12 UAVs, the controller is presented as achieving safe formation change, passage through building gaps, avoidance of static and dynamic obstacles, and formation recovery, with inter-drone distances held above the safety threshold.

Load-bearing premise

The load-bearing premise is that rotating the obstacle-relative displacement inside the avoidance potential changes the potential's value and force direction enough to add 3D maneuverability; if the rotation leaves the norm unchanged, the avoidance term collapses to radial repulsion and the distinctive 3D claim loses its mechanism.

Editorial extensions

If this is right

  • Multi-UAV swarms could perform formation changes and recovery in cluttered urban areas without consensus-based coordination, lowering controller design complexity.
  • The centralized/distributed split scales to at least 12 UAVs in the presented simulations, with only local neighbor and obstacle information used by the safety layer.
  • Dynamic obstacles can be handled online by range and field-of-view detection, rather than by precomputed global path planning.
  • The same controller structure could be reused for non-static barrier areas, since the Lloyd targets are recomputed as the barrier moves.

Reading between the lines

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

  • Editorially, the 3D claim stands or falls on whether the rotation matrices change the force direction; because rotation preserves vector norms, the potential as written is radial, so a genuine 3D maneuver needs an additional mechanism such as heading-dependent asymmetry in the potential.
  • A testable extension is to replace the rotational potential with an explicit tangential steering term and compare minimum-distance statistics against the present formulation in the same obstacle layouts.
  • The modular architecture suggests a practical deployment principle: formation targets can update at a slower global rate while avoidance runs at a fast local rate; a hardware or high-fidelity simulator test could measure how much update-rate separation the safety distances tolerate.
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

3 major / 5 minor

Summary. This paper proposes a hybrid control framework for multi-UAV formation flight: a centralized probabilistic Lloyd's algorithm determines optimal formation positions, a distributed Hooke's-law term handles inter-vehicle collision avoidance, and a pigeon-inspired potential field is used for obstacle avoidance, with a claimed extension to 3D via rotational potentials. The method is evaluated in three simulation case studies (8 and 12 UAVs) involving static and dynamic obstacles, buildings, formation changes, and formation recovery.

Significance. If correct, the framework would offer a relatively simple, non-consensus-based alternative for scalable multi-UAV formation control with 3D obstacle avoidance. However, the central mathematical contribution is not sound: the 3D rotational potential is invariant under the rotation matrices used, so it reduces to an ordinary radial repulsion, and the printed gradient formulas are not actual gradients. This collapses the claimed 3D maneuverability, which is the paper's primary novelty. The Lloyd-based formation controller and the inter-vehicle collision term may still be viable, but they are not the claimed contribution.

major comments (3)
  1. [2.4.2 (Eqs. 20-24)] The rotational potential U_r in Eq. (20) is invariant under the rotation matrices T_rx, T_ry, T_rz defined in Eqs. (21)-(23), because the product of rotation matrices is orthogonal and thus preserves the Euclidean norm. Hence U_r(p_i,o_k) = (k_r/2)||p_i-o_k||^2 for every alpha, and the angle alpha in Eq. (16) cancels identically. Consequently, the claimed 'additional maneuverability' and the 3D steering force do not exist in the model as written. Furthermore, Eq. (24) is not the gradient of Eq. (20): differentiating Eq. (20) gives k_r(p_i-o_k) (with additional alpha-dependent terms if alpha is treated as position-dependent), whereas Eq. (24) is a rotated radial vector field that generally has nonzero curl and therefore cannot be the gradient of any scalar potential. This invalidates the central contribution of the paper.
  2. [2.4.1 (Eqs. 11-12)] The gradient of U_p given in Eq. (12) is incorrect for a non-identity diagonal matrix k_v. The chain rule applied to U_p = (1/2)(||k_v(p_i-o_k)|| - r_a)^2 yields ∇U_p = (||k_v(p_i-o_k)|| - r_a) * (k_v^T k_v(p_i-o_k)) / ||k_v(p_i-o_k)||, not the expression with ∇(||p_i-o_k||) as printed. As a consequence, the obstacle avoidance control law in Eq. (17) does not implement gradient descent on the declared potential, and the 'potential field' interpretation fails. Since Eq. (17) is also used in the 3D scenario, this error propagates into the main simulations.
  3. [2.5 (Eq. 25)] The overall control law in Eq. (25) is formed by superposing the formation, collision, and obstacle avoidance terms, and the text claims stability from the 'superposition principle,' but no Lyapunov, invariance, or barrier-function argument is provided to show that the closed-loop system is collision-free in the presence of moving obstacles. The only evidence is the distance histories in Figures 5-7, which are qualitative and do not substantiate the 'collision-free' guarantee stated in the title and abstract.
minor comments (5)
  1. [Section 2 (Eqs. 1-2)] Equation (2) is used twice with the same number: once for the velocity dynamics ḍ_i = u_i and once for the neighborhood definition N_i = {j : ||p_i-p_j|| < r_d}.
  2. [Section 2 (Eq. 5)] In the sentence preceding Eq. (5), the cost function being differentiated is defined in Eq. (3), not Eq. (4); Eq. (4) defines the Voronoi partition.
  3. [Eq. (10)] The planar detection condition in Eq. (10) includes a term with z-coordinates (atan((o_kz-p_iz)/(o_kx-p_ix))), which is inconsistent with the stated planar 2D setting.
  4. [Table 2] Table 2 lists r_ok in units of m/s, but r_ok is used as an obstacle radius (a length); also, the dynamic obstacle velocity in Case Study 2 (v_obs4 = [0.1,0.025,0]^T m/s) differs from the value in Table 2 ([0.2,0.05,0]).
  5. [References] Reference [25] contains a typo in the title ('Dyamics' should be 'Dynamics'), and the text in Section 2.3 refers to a 'mass-spiring-damper system' rather than 'mass-spring-damper.'

Circularity Check

1 steps flagged · score 6.0 of 10

The central 3D avoidance potential is rotation-invariant: Eq. (20) equals (k_r/2)||p_i-o_k||^2, so the claimed 3D maneuver reduces to a standard distance repulsion.

  1. renaming known result [Section 2.4.2 'Nonplanar Velocity Adjustment (3D)', Eqs. (20)-(24), with alpha from Eq. (16)]
    "To extend this pigeon-inspired maneuver into the 3D space, the rotational potential function presented in Eq. (13) needs to be reformulated as below: U_r(p_i,o_k)= { (k_r/2)(||T_rx T_ry T_rz(p_i-o_k)||)^2, detected; 0, otherwise (20) ... T_rx = [1 0 0; 0 1 0; 0 0 1] (21), T_ry = [cos alpha 0 sin alpha; 0 1 0; -sin alpha 0 cos alpha] (22), T_rz = [cos alpha -sin alpha 0; sin alpha cos alpha 0; 0 0 1] (23)."

    T_rx, T_ry, and T_rz are rotation matrices, so their product is orthogonal and norm-preserving: for every vector v, ||T_rx T_ry T_rz v|| = ||v||. Substituting v = p_i - o_k makes Eq. (20) identically equal to (k_r/2)||p_i-o_k||^2, independent of the piecewise angle alpha defined in Eq. (16). Thus the '3D rotational potential' is exactly the standard distance-based repulsion already used in the planar controller; it cannot supply 'additional maneuverability' or any alpha-dependent 3D steering force. Eq. (24) is not the correct gradient of Eq. (20), and even a correct gradient would contain no alpha-dependence because alpha cancels from the potential. The claimed 3D maneuver is therefore equivalent by construction to the input repulsion term, merely relabeled with rotation matrices.

full rationale

The formation-control part (probabilistic Lloyd/CVT) and the inter-vehicle collision-avoidance term are not circular: the Lloyd algorithm is an external computational tool and the collision term follows a standard potential-force structure. The circular/vacuous step is confined to the advertised main novelty, the 3D obstacle-avoidance maneuver. The paper defines a 'rotational potential' in Eq. (20) using rotation matrices T_rx, T_ry, T_rz; because rotations preserve Euclidean norms, this potential is literally the isotropic squared-distance repulsion (k_r/2)||p_i-o_k||^2, with the angle alpha from Eq. (16) cancelling identically. Consequently, the mathematical mechanism claimed as the source of 3D maneuvers is absent from the model as written. The printed gradient Eq. (24) is not a derivative of Eq. (20) and does not reintroduce alpha-dependence. This makes the central 3D contribution a relabeling of a known repulsion rather than a derived 3D avoidance law. The simulations do not repair the derivation because they implement the posited equations. Separately, Eq. (12) also mis-differentiates Eq. (11) when k_v is not isotropic, but that is a correctness issue rather than circularity. Since the central advertised contribution reduces by construction while the formation and collision components remain independent, the appropriate score is 6.

Assumptions & free parameters 9 free parameters · 5 assumptions · 0 invented entities

The central claim rests on hand-tuned controller gains, an unsupported superposition assumption, and a mathematically invalid rotational potential. No code, data, or baseline comparisons are provided.

free parameters (9)
  • K_p = diag([3,3,3])
    Formation position gain; paper says 'can be tuned through trial and error' (Section 2.3).
  • K_v = diag([5,5,5])
    Formation velocity gain; tuned by trial and error.
  • k_v = diag([0.1,0.5,0.1])
    Anisotropic obstacle-distance matrix in Eq. (11); dominates the direction of the avoidance force, set by hand.
  • k_r = 0.5
    Gain of the 'rotational' potential, set by hand.
  • k_o1 = 5
    Obstacle repulsion gain.
  • k_o2 = 1
    Obstacle damping gain.
  • k_c1, k_c2 = not specified
    Collision avoidance gains in Eq. (7); no values appear in Table 2, blocking reproduction.
  • alpha_1, alpha_2, beta_1, beta_2 = not specified
    Lloyd's algorithm constants in Table 1 must satisfy alpha1+alpha2=1, beta1+beta2=1; numerical values are never given.
  • r_d, r_s, r_ok, theta_FOV = 2, 1, 1, 60 deg
    Detection range, safety range, obstacle radius, and FOV angle; chosen scenario parameters.
assumptions (5)
  • domain assumption UAVs are modeled as point masses with double-integrator dynamics (Eqs. 1-2).
    Stated in Section 2.1; ignores attitude, actuator, and aerodynamic limits that matter in dense urban flight.
  • ad hoc to paper The total control input is the sum of formation, collision, and obstacle terms (Eq. 25), and stability follows from superposition.
    No stability or convergence proof is provided; the superposition is asserted.
  • standard math Rotation matrices preserve Euclidean norms.
    Implied by the definition of U_r in Eqs. (13) and (20); the paper's gradient formulas in Eqs. (14) and (24) contradict this fact.
  • domain assumption Probabilistic Lloyd's algorithm converges to a centroidal Voronoi configuration.
    Taken from [19] without reproducing a convergence proof or specifying constants.
  • domain assumption Obstacles are detected as points with known position and radius, and buildings are static obstacles represented only informally in figures.
    The detection conditions in Eqs. (9)-(10), (18)-(19) treat obstacles as ideal points; building geometry is not formally defined.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Swarm Intelligence in Collision-free Formation Control for Multi-UAV Systems with 3D Obstacle Avoidance Maneuvers." pith.science (2026). https://pith.science/paper/HRSDJXUQ

@misc{pith2026241212437,
  author       = {Pith},
  title        = {Pith review of: Swarm Intelligence in Collision-free Formation Control for Multi-UAV Systems with 3D Obstacle Avoidance Maneuvers},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HRSDJXUQ}},
  note         = {Machine review of arXiv:2412.12437}
}
read the original abstract

Recent advances in multi-agent systems manipulation have demonstrated a rising demand for the implementation of multi-UAV systems in urban areas which are always subjected to the presence of static and dynamic obstacles. The focus of the presented research is on the introduction of a nature-inspired collision-free control for a multi-UAV system considering obstacle avoidance maneuvers. Inspired by the collective behavior of tilapia fish and pigeon, the presented framework in this study uses a centralized controller for the optimal formation control/recovery, which is defined by probabilistic Lloyd's algorithm, while it uses a distributed controller for the intervehicle collision and obstacle avoidance. Further, the presented framework has been extended to the 3D space with 3D maneuvers. Finally, the presented framework has been applied to a multi-UAV system in 2D and 3D scenarios, and obtained results demonstrated the validity of the presented method in the presence of buildings and different types of obstacles.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Neuromorphic Digital-Twin-based Controller for Indoor Multi-UAV Systems Deployment

    cs.NE 2025-02 conditional novelty 4.0 of 10

    A cloud-edge framework in which each UAV runs a spiking neural network that learns to mimic a cloud-based nature-inspired controller, cutting computational load by roughly 90% in simulations.

Reference graph

Works this paper leans on

25 extracted references · 9 canonical work pages · cited by 1 Pith paper

  1. [1]

    A review on representative swarm intelligence algorithms for solving optimization problems: Applications and trends,

    J. Tang, G. Liu and Q. Pan , "A review on representative swarm intelligence algorithms for solving optimization problems: Applications and trends," IEEE/CAA Journal of Automatica Sinica, pp. 1627-1643, 2021

    2021

  2. [2]

    Drone swarm strategy for the detection and tracking of occluded targets in complex environments,

    R. J. Amala Arokin Nathan , I. Kurmi and O. Bimber , "Drone swarm strategy for the detection and tracking of occluded targets in complex environments," Communications Engineering, vol. 2, no. 1, p. 55, 2023

    2023

  3. [3]

    Distributed cooperative obstacle avoidance and formation reconfiguration for multiple quadrotors: Theory and experiment,

    J. Guo, J. Qi, M. Wang, C. Wu, Y. Ping, S. Li and J. Jin, "Distributed cooperative obstacle avoidance and formation reconfiguration for multiple quadrotors: Theory and experiment," Aerospace Science and Technology, vol. 136, p. 108218, 2023

    2023

  4. [4]

    Computationally Efficient Visibility Graph-Based Generation Of 3D Shortest Collision-Free Path Among Polyhedral Obstacles For Unmanned Aerial Vehicles,

    S. Huang and R. S. H. Teo, "Computationally Efficient Visibility Graph-Based Generation Of 3D Shortest Collision-Free Path Among Polyhedral Obstacles For Unmanned Aerial Vehicles," in 2019 International Conference on Unmanned Aircraft Systems (ICUAS), 2019, (pp. 1218-1223)

    work page 2019

  5. [5]

    Selective velocity obstacle method for deconflicting maneuvers applied to unmanned aerial vehicles,

    Y. I. Jenie, E. J. Kampen, C. C. de Visser, J. Ellerborek and J. M. Hoekstra, "Selective velocity obstacle method for deconflicting maneuvers applied to unmanned aerial vehicles," Journal of Guidance, Control, and Dynamics, vol. 38, no. 6, pp. 1140-1146, 2015

    2015

  6. [6]

    Modified artificial potential field method for online path planning applications,

    F. Bounini, D. Gingras, H. Pollart and D. Gruyer, "Modified artificial potential field method for online path planning applications," in 2017 IEEE Intelligent Vehicles Symposium (IV), (pp. 180-185), 2017

    2017

  7. [7]

    Predictive control of aerial swarms in cluttered environments,

    E. Soria, F. Schiano and D. Floreano, "Predictive control of aerial swarms in cluttered environments," Nature Machine Intelligence, vol. 3, no. 6, pp. 545-554, 2021

    2021

  8. [8]

    A potential field -based model predictive path -planning controller for autonomous road vehicles,

    Y. Rasekhipour, A. Khajepour, S. K. Chen and B. Litkouhi, "A potential field -based model predictive path -planning controller for autonomous road vehicles," IEEE Transactions on Intelligent Transportation Systems, vol. 18, no. 5, pp. 1255-1267, 2016

    work page 2016

Show all 25 references
  1. [9]

    A Novel Hybrid Particle Swarm Optimization Algorithm for Path Planning of UAVs,

    Z. Yu, Z. Si, X. Li, D. Wang and H. Song, "A Novel Hybrid Particle Swarm Optimization Algorithm for Path Planning of UAVs," IEEE Internet of Things Journal, vol. 9, no. 22, pp. 22547-22558, 2021

    2021

  2. [10]

    Chaos Particle Swarm Optimization Enhancement Algorithm for UAV Safe Path Planning,

    H. Chu, J. Yi and F. Yang, "Chaos Particle Swarm Optimization Enhancement Algorithm for UAV Safe Path Planning," Applied Sciences, vol. 12, no. 18, p. 8977, 2022

    2022

  3. [11]

    Path Planning Method With Improved Artificial Potential Field—A Reinforcement Learning Perspective,

    Q. Yao, Z. Zheng, L. Qi, H. Yuan, X. Guo, M. Zhao, Z. Liu and T. Yang , "Path Planning Method With Improved Artificial Potential Field—A Reinforcement Learning Perspective," IEEE Access, vol. 8, pp. 135513-135523, 2020

    2020

  4. [12]

    An Improved Artificial Potential Field Method for Path Planning and Formation Control of the Multi -UAV Systems,

    Z. Pan, C. Zhang, Y. Xia, H. Xiong and X. Shao, "An Improved Artificial Potential Field Method for Path Planning and Formation Control of the Multi -UAV Systems," IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 69, no. 3, pp. 1129-1133, 2022

    2022

  5. [13]

    Uav path planning based on improved artificial potential field method,

    G. Hao, Q. Lv, Z. Huang, H. Zhao and W. Chen, "Uav path planning based on improved artificial potential field method," Aerospace, vol. 10, no. 6, p. 562, 2023

    2023

  6. [14]

    B -APFDQN: A UAV path planning algorithm based on deep Q -network and artificial potential field,

    F. Kong, Q. Wang, S. Gao and H. Yu, "B -APFDQN: A UAV path planning algorithm based on deep Q -network and artificial potential field," IEEE Access, vol. 11, pp. 44051-44064, 2023

    2023

  7. [15]

    Through the eyes of a bird: modelling visually guided obstacle flight,

    H. T. Lin, I. G. Ros and A. A. Biewener, "Through the eyes of a bird: modelling visually guided obstacle flight," Journal of the Royal Society Interface, vol. 11, no. 96, p. 20140239, 2014

    2014

  8. [16]

    Live -fly experimentation for pigeon-inspired obstacle avoidance of quadrotor unmanned aerial vehicles,

    M. Huo, H. Duan, Q. Yang, D. Zhang and H. Qiu, "Live -fly experimentation for pigeon-inspired obstacle avoidance of quadrotor unmanned aerial vehicles," Science China Information Sciences, vol. 62, pp. 1-8, 2019

    2019

  9. [17]

    Multiobjective Spherical Vector-Based Particle Swarm Optimisation for 3D UAV Path Planning,

    M. Zhang, K. Ma, L. Liu and W. Zhang, "Multiobjective Spherical Vector-Based Particle Swarm Optimisation for 3D UAV Path Planning," in 2024 6th International Conference on Data -driven Optimization of Complex Systems (DOCS) , (pp. 427-432), 2024

    2024

  10. [18]

    A Multiobjective Optimization Algorithm for Safety and Optimality of 3-D Route Planning in UAV,

    M. Abdel -Basset, R. Mohamed, K. M. Sallam, I. M. Hezam, K. Munasinghe and A. Jamalipour, "A Multiobjective Optimization Algorithm for Safety and Optimality of 3-D Route Planning in UAV," IEEE Transactions on Aerospace and Electronic Systems, 2024

    2024

  11. [19]

    Voronoi tessellations and their application to climate and global modeling,

    L. Ju, T. Ringler and M. Gunzburger, "Voronoi tessellations and their application to climate and global modeling," in Numerical techniques for global atmospheric models, Springer, 2011, pp. 313-342

    2011

  12. [20]

    Computationally-efficient distributed algorithms of navigation of teams of autonomous UAVs for 3D coverage and flocking,

    T. Elmokadem and A. V. Savkin, "Computationally-efficient distributed algorithms of navigation of teams of autonomous UAVs for 3D coverage and flocking," Drones, vol. 5, no. 4, p. 124, 2021

    2021

  13. [21]

    Coverage control for mobile sensing networks,

    J. Cortes, S. Martinez, T. Karatas and F. Bullo, "Coverage control for mobile sensing networks," IEEE Transactions on robotics and Automation, vol. 20, no. 2, pp. 243-255, 2004

    2004

  14. [22]

    Flocking for multi -agent dynamic systems: Algorithms and theory,

    R. Olafti-Saber, "Flocking for multi -agent dynamic systems: Algorithms and theory," IEEE Transactions on automatic control, vol. 51, no. 3, pp. 401-420, 2006

    2006

  15. [23]

    Spatially -distributed coverage optimization and control with limited -range interactions,

    A. Cortes, S. Martinez and F. Bullo, "Spatially -distributed coverage optimization and control with limited -range interactions," ESAIM: Control, Optimisation and Calculus of Variations, vol. 11, no. 4, pp. 691-719, 2005

    2005

  16. [24]

    Hexagonal territories,

    G. W. Barlow, "Hexagonal territories," Animal Behavior, vol. 22, pp. 876-IN1, 1974

    1974

  17. [25]

    Attitude Dynamics and Control,

    B. Wie, "Attitude Dynamics and Control," in Space Vehicle Dyamics and Control , Amaerican Institute of Aeronautics and Astronautics, 1998, pp. 307-322

    1998

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.