pith. sign in

arxiv: 2604.05677 · v1 · submitted 2026-04-07 · 💻 cs.RO

Dynamic Control Allocation for Dual-Tilt UAV Platforms

Marcello Sorge , Federico Ciresola , Giulia Michieletto , Angelo Cenedese This is my paper

Pith reviewed 2026-05-10 19:15 UTC · model grok-4.3

classification 💻 cs.RO
keywords control allocationdual-tilt UAVhexarotoractuator saturationtrajectory trackinghierarchical control
0
0 comments X

The pith

A dynamic allocator for dual-tilt UAVs imposes first-order actuator dynamics and optimizes states while modeling saturation effects.

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

The paper develops a two-layer controller for a hexarotor UAV whose propellers can each tilt independently along two axes. A high-level module computes the total wrench needed to track a trajectory. A lower-level allocator then solves for motor speeds and tilt angles that deliver this wrench, forcing the actuators to follow chosen first-order responses and using extra degrees of freedom to minimize a user-defined cost. The allocator treats saturation limits explicitly and supplies analysis of how those limits alter tracking accuracy. Simulations confirm the scheme succeeds for path-following tasks.

Core claim

We present a hierarchical control structure composed of a high-level controller generating the required wrench for the tracking task, and a control allocation law ensuring that the actuators produce such wrench. The allocator imposes desired first-order dynamics on the actuators set, and exploits system redundancy to optimize the actuators state with respect to a given objective function. Unlike other studies on the subject, we explicitly model actuator saturation and provide theoretical insights on its effect on control performances. We also investigate the role of propeller tilt angles, by imposing asymmetric shapes in the objective function.

What carries the argument

The dynamic control allocation law that enforces first-order actuator dynamics while solving a redundancy-optimization problem subject to saturation constraints.

Load-bearing premise

The UAV platform permits each propeller to tilt independently along two orthogonal axes throughout flight.

What would settle it

A numerical or physical test in which commanded wrench cannot be produced once saturation occurs, causing measured tracking error to exceed the value predicted by the unsaturated allocator equations.

Figures

Figures reproduced from arXiv: 2604.05677 by Angelo Cenedese, Federico Ciresola, Giulia Michieletto, Marcello Sorge.

Figure 1
Figure 1. Figure 1: Hierarchical control architecture showing the high-level controller and the low-level allocator [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 1
Figure 1. Figure 1: Remark 1. Asymptotic convergence of uv to u ‹ v is not guar￾anteed if the actuators state variables reach their saturation limits, due to the matrix ∇satpxaq not being invertible. This can happen if the desired trajectory is too fast for the actuation capabilities of the robot, or if the high-level controller is too aggressively tuned. In the following, the operator ∇satpxaq is computed as ∇satpxa,hq “ # 1… view at source ↗
Figure 4
Figure 4. Figure 4: Objective functions 0 10 20 30 40 -10 0 10 20 ,1 [/ ] 0 10 20 30 40 -20 -10 0 10 ,2 [/ ] 0 10 20 30 40 -20 -10 0 10 ,3 [/ ] 0 10 20 30 40 -20 -10 0 10 ,4 [/ ] 0 10 20 30 40 time [s] -10 0 10 20 ,5 [/ ] 0 10 20 30 40 time [s] -10 0 10 20 ,6 [/ ] .j = 0 .j = 10 [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Propeller α angles 0 10 20 30 40 -10 0 10 20 -1 [/ ] 0 10 20 30 40 -10 0 10 20 -2 [/ ] 0 10 20 30 40 -10 0 10 20 -3 [/ ] 0 10 20 30 40 -20 -10 0 10 -4 [/ ] 0 10 20 30 40 time [s] -20 -10 0 10 -5 [/ ] 0 10 20 30 40 time [s] -20 -10 0 10 -6 [/ ] .j = 0 .j = 10 [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 7
Figure 7. Figure 7: Propeller spinning rates show an anti-symmetric behavior. Moreover, it is evident that the tilt angles oscillate around the middle of the saturation intervals. From figure 7, it can be notice how the magnitude of the propeller spinning rates reduces when the term uj is [PITH_FULL_IMAGE:figures/full_fig_p005_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Propeller [PITH_FULL_IMAGE:figures/full_fig_p007_8.png] view at source ↗
Figure 11
Figure 11. Figure 11: Propeller [PITH_FULL_IMAGE:figures/full_fig_p007_11.png] view at source ↗
read the original abstract

This paper focuses on dynamic control allocation for a hexarotor UAV platform, considering a trajectory tracking task as as case study. It is assumed that the platform is dual-tilting, meaning that it is able to tilt each propeller independently during flight, along two orthogonal axis. We present a hierarchical control structure composed of a high-level controller generating the required wrench for the tracking task, and a control allocation law ensuring that the actuators produce such wrench. The allocator imposes desired first-order dynamics on the actuators set, and exploits system redundancy to optimize the actuators state with respect to a given objective function. Unlike other studies on the subject, we explicitly model actuator saturation and provide theoretical insights on its effect on control performances. We also investigate the role of propeller tilt angles, by imposing asymmetric shapes in the objective function. Numerical simulations are presented to validate the allocation strategy.

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 / 3 minor

Summary. The paper proposes a hierarchical control architecture for trajectory tracking on a dual-tilt hexarotor UAV. A high-level controller generates the required wrench; a dynamic allocator then maps the wrench to actuator commands while enforcing first-order actuator dynamics, exploiting redundancy to minimize a user-specified objective function, and explicitly incorporating actuator saturation limits. Theoretical discussion of saturation effects is provided, asymmetric tilt-angle effects are explored by shaping the objective function, and the allocator is validated in numerical simulation.

Significance. If the saturation analysis and simulation results hold, the work supplies a practical, dynamic allocation method for over-actuated tilting-rotor platforms that directly accounts for actuator limits rather than treating them post hoc. The combination of imposed first-order dynamics with redundancy optimization is standard in control allocation but is here specialized to dual-tilt geometry; explicit saturation modeling and the asymmetric-objective investigation are incremental but useful contributions to UAV control literature.

major comments (2)
  1. [§4, §5] §4 (Control Allocation Law) and §5 (Saturation Analysis): the claimed 'theoretical insights' on saturation effects are not accompanied by explicit bounds, stability margins, or a formal proof that the allocator remains well-posed when saturation is active; the section appears to rely on qualitative discussion and simulation observations rather than load-bearing derivations.
  2. [§6] §6 (Numerical Simulations): no quantitative performance metrics (RMS tracking error, actuator effort, saturation frequency), no error bars or Monte-Carlo statistics, and no direct comparison against a baseline allocator that ignores saturation are reported; this weakens the validation of the central claim that explicit saturation modeling improves performance.
minor comments (3)
  1. [Abstract] Abstract: 'as as case study' is a typographical error.
  2. [§4] The objective-function weights are listed as free parameters; a brief sensitivity study or selection guideline would improve reproducibility.
  3. [§2, §3] Notation for the wrench vector and actuator mapping matrix should be introduced once and used consistently across sections.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback and the recommendation for minor revision. We address the major comments point by point below, indicating the revisions we will make to improve the clarity and rigor of the theoretical discussion and simulation validation.

read point-by-point responses

  1. Referee: [§4, §5] §4 (Control Allocation Law) and §5 (Saturation Analysis): the claimed 'theoretical insights' on saturation effects are not accompanied by explicit bounds, stability margins, or a formal proof that the allocator remains well-posed when saturation is active; the section appears to rely on qualitative discussion and simulation observations rather than load-bearing derivations.

    Authors: We agree that the saturation analysis in §5 is primarily qualitative, drawing on the structure of the dynamic allocator and observations from simulation to illustrate effects on wrench tracking and actuator commands. The manuscript does not derive explicit bounds on saturation-induced errors or provide a formal proof of well-posedness under active saturation. In the revision we will expand §5 with a more precise characterization of the feasible set under saturation (including conditions for existence of solutions to the constrained optimization) and clarify the scope of the claimed insights. A full closed-loop stability analysis with saturation is beyond the paper's focus on the allocation layer and would require additional assumptions on the high-level controller; we will note this limitation explicitly. revision: partial

  2. Referee: [§6] §6 (Numerical Simulations): no quantitative performance metrics (RMS tracking error, actuator effort, saturation frequency), no error bars or Monte-Carlo statistics, and no direct comparison against a baseline allocator that ignores saturation are reported; this weakens the validation of the central claim that explicit saturation modeling improves performance.

    Authors: We accept that the simulation results would be strengthened by quantitative metrics and comparisons. In the revised manuscript we will report RMS position and attitude tracking errors, integrated actuator effort, and saturation frequency for the proposed allocator. We will also include a direct comparison against a baseline dynamic allocator that does not incorporate saturation constraints (e.g., by clipping or ignoring limits), and we will add Monte-Carlo statistics over multiple initial conditions or disturbance realizations where computationally feasible. These additions will provide clearer evidence for the performance benefit of explicit saturation modeling. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation is self-contained design

full rationale

The paper defines a hierarchical controller for dual-tilt hexarotor trajectory tracking: a high-level wrench generator followed by an allocator that imposes first-order actuator dynamics, exploits redundancy through an objective function, and explicitly incorporates saturation bounds. These elements are presented as explicit design choices whose performance consequences are then analyzed theoretically and via simulation. The dual-tilt capability is stated upfront as the platform premise rather than derived. No equation reduces by construction to a fitted parameter, no prediction is statistically forced from its own inputs, and no load-bearing uniqueness or ansatz is imported via self-citation. The central claims therefore remain independent of the inputs they act upon.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Based solely on abstract; limited visibility into parameters or assumptions. The dual-tilt capability is treated as given, and the objective function likely involves tunable weights for asymmetry.

free parameters (1)
  • objective function weights
    Asymmetric shapes imposed on the objective function to investigate tilt angle roles; these are chosen parameters affecting optimization.
axioms (1)
  • domain assumption The UAV platform allows independent dual tilting of each propeller along two orthogonal axes.
    Explicitly stated as an assumption enabling the control allocation and redundancy exploitation.

pith-pipeline@v0.9.0 · 5446 in / 1357 out tokens · 60703 ms · 2026-05-10T19:15:12.515417+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

10 extracted references · 10 canonical work pages · 1 internal anchor

  1. [1]

    Hierarchical control of the over-actuated ROSPO platform via static input allocation

    Nainer, C., et al. "Hierarchical control of the over-actuated ROSPO platform via static input allocation." IFAC-PapersOnLine 50.1 (2017): 12698-12703

    work page 2017

  2. [2]

    Optimality-based dynamic allocation with nonlinear first-order redundant actuators

    Passenbrunner, Thomas E., Mario Sassano, and Luca Zaccarian. "Optimality-based dynamic allocation with nonlinear first-order redundant actuators." European Journal of Control 31 (2016): 33-40

    work page 2016

  3. [3]

    Input allocation for the propeller-based overactuated platform ROSPO

    Furci, Michele, et al. "Input allocation for the propeller-based overactuated platform ROSPO." IEEE Transactions on Control Systems Technology 28.6 (2019): 2720-2727

    work page 2019

  4. [4]

    Towards spline-based dynamic input allocation

    Akbari, Shima, Sergio Galeani, and Mario Sassano. "Towards spline-based dynamic input allocation." 2023 International Conference on Control, Automation and Diagnosis (ICCAD). IEEE, 2023

    work page 2023

  5. [5]

    Spline-based Input Allocation On An Overactuated Tilting and Twisting Unmanned Aerial Vehicle

    Akbari, Shima, et al. "Spline-based Input Allocation On An Overactuated Tilting and Twisting Unmanned Aerial Vehicle." 2024 10th International Conference on Control, Decision and Information Technologies (CoDIT). IEEE, 2024

    work page 2024

  6. [6]

    Output invisible control allocation with asymptotic optimization for nonlinear systems in normal form

    Cristofaro, Andrea, Sergio Galeani, and Andrea Serrani. "Output invisible control allocation with asymptotic optimization for nonlinear systems in normal form." 2017 IEEE 56th Annual Conference on Decision and Control (CDC). IEEE, 2017

    work page 2017

  7. [7]

    Dynamic allocation for input redundant control systems

    Zaccarian, Luca. "Dynamic allocation for input redundant control systems." Automatica 45.6 (2009): 1431-1438

    work page 2009

  8. [8]

    Design and optimal control of a tiltrotor micro-aerial vehicle for efficient omnidirectional flight

    Allenspach, Mike, et al. "Design and optimal control of a tiltrotor micro-aerial vehicle for efficient omnidirectional flight." The International Journal of Robotics Research 39.10-11 (2020): 1305-1325

    work page 2020

  9. [9]

    Allocation for Omnidirectional Aerial Robots: Incorporating Power Dynamics

    Cuniato, Eugenio, et al. "Allocation for Omnidirectional Aerial Robots: Incorporating Power Dynamics." arXiv preprint arXiv:2412.16107 (2024)

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  10. [10]

    DӺAxpõ *c \< Tq*(n º _ = bX ux am ̵ PJ T! 4? *dϸ z ?ȒP d 3 ʍ ]\

    11em plus .33em minus .07em 4000 4000 100 4000 4000 500 `\.=1000 = #1 \@IEEEnotcompsoconly \@IEEEcompsoconly #1 * [1] 0pt [0pt][0pt] #1 * [1] 0pt [0pt][0pt] #1 * \| ** #1 \@IEEEauthorblockNstyle \@IEEEcompsocnotconfonly \@IEEEauthorblockAstyle \@IEEEcompsocnotconfonly \@IEEEcompsocconfonly \@IEEEauthordefaulttextstyle \@IEEEcompsocnotconfonly \@IEEEauthor...

    work page arXiv 2052