2025-11-12T03:49:09.545098

Trajectory control of a suspended load with non-stopping flying carriers

Girardello, Michieletto, Cenedese et al.
This paper presents the first closed-loop control framework for cooperative payload transportation with non-stopping flying carriers. Building upon grasp-matrix formulations and internal force redundancy, we propose a feedback wrench controller that actively regulates the payload's pose while an optimization layer dynamically shapes internal-force oscillations to guarantee persistent carrier motion. Preliminary experimental results on multirotor UAVs validate the model assumptions, and numerical simulations demonstrate that the method successfully prevents carrier stagnation, achieves accurate load tracking, and generates physically feasible trajectories with smooth velocity profiles. The proposed framework not only advances the state of the art but also offers a reliable, versatile solution for future real-world applications requiring load transportation by coordinated non-stopping flying carriers.
academic

Trajectory control of a suspended load with non-stopping flying carriers

Basic Information

  • Paper ID: 2510.11413
  • Title: Trajectory control of a suspended load with non-stopping flying carriers
  • Authors: Sofia Girardello, Giulia Michieletto, Angelo Cenedese, Antonio Franchi, Chiara Gabellieri
  • Classification: eess.SY cs.SY (Systems and Control)
  • Publication Date: October 13, 2025 (arXiv preprint)
  • Paper Link: https://arxiv.org/abs/2510.11413

Abstract

This paper presents the first closed-loop control framework for cooperative load transportation using non-stopping flying carriers. Based on grasp matrix formulation and internal force redundancy, a feedback torque controller is proposed to actively regulate load attitude, while an optimization layer dynamically adjusts internal force oscillations to ensure continuous carrier motion. Preliminary experimental results on multirotor unmanned aerial vehicles validate the model assumptions, and numerical simulations demonstrate that the method successfully prevents carrier stagnation, achieves precise load tracking, and generates physically feasible trajectories with smooth velocity profiles.

Research Background and Motivation

Research Problem

The core problem addressed in this paper is: how to achieve closed-loop control for cooperative transportation of suspended loads using non-stopping flying carriers (such as fixed-wing unmanned aerial vehicles), particularly when the load requires dynamic trajectory tracking.

Problem Significance

  1. Application Demand: Long-distance logistics, humanitarian relief, infrastructure inspection, and construction applications require urgent load transportation capabilities
  2. Technical Challenge: Fixed-wing unmanned aerial vehicles offer superior range and endurance compared to multirotor platforms but must maintain forward velocity to sustain flight
  3. Practical Significance: Provides theoretical foundation and technical solutions for future large-scale, long-distance cooperative aerial transportation

Limitations of Existing Methods

  1. Multirotor Constraints: Existing research primarily focuses on multirotor unmanned aerial vehicles, which have limited endurance and operational range
  2. Open-Loop Control: Existing non-stopping carrier research is limited to open-loop trajectory planning, lacking closed-loop feedback control
  3. Static Loads: Current methods only consider static load equilibrium without supporting dynamic trajectory tracking

Core Contributions

  1. First Closed-Loop Control Framework: Proposes the first closed-loop control system for cooperative load transportation using non-stopping flying carriers
  2. Dynamic Trajectory Tracking: Extends existing theory to support dynamic load trajectory tracking rather than merely static equilibrium
  3. Online Optimization Strategy: Develops real-time optimization algorithms that dynamically adjust internal force parameters to ensure carrier velocity constraints
  4. Theoretical and Experimental Validation: Verifies method effectiveness through numerical simulation and preliminary experiments

Methodology Details

Task Definition

Input: Desired load trajectory pLd(t)∈R3p_{Ld}(t) \in \mathbb{R}^3, RLd(t)∈SO(3)R_{Ld}(t) \in SO(3), and corresponding velocities Output: Cable force trajectories {fi(t)}i=1n\{f_i(t)\}_{i=1}^n and carrier trajectories {pRi(t)}i=1n\{p_{R_i}(t)\}_{i=1}^nConstraints:

  1. Load tracking error is asymptotically stable
  2. Carrier velocity constraint: ∥p˙Ri(t)∥≥ε>0\|\dot{p}_{R_i}(t)\| \geq \varepsilon > 0, ∀i=1,...,n\forall i = 1,...,n

Model Architecture

1. System Modeling

The system comprises n≥3n \geq 3 non-stopping flying carriers, each connected to a common rigid load via cables. Key kinematic relationships:

pRi=pL+RLBbi+Liqi,i=1,...,np_{R_i} = p_L + R_L {}^B b_i + L_i q_i, \quad i = 1,...,n

where qi∈S2q_i \in S^2 is the unit direction vector from load to carrier.

2. Grasp Matrix Representation

The relationship between cable forces and load torques is described through the grasp matrix:

w=G(RL)fw = G(R_L)f

where G(RL)∈R6×3nG(R_L) \in \mathbb{R}^{6 \times 3n} is the grasp matrix:

I_3 & \cdots & I_3 \\ S({}^B b_1)R_L^T & \cdots & S({}^B b_n)R_L^T \end{bmatrix}$$ #### 3. Three-Layer Control Architecture **Outer Loop Load Torque Controller**: ``` fd = -Kp ep - Kv ev - Ki ∫₀ᵗ ep(τ)dτ + mL ge₃ τd = -KR eR - Kω eω - KiR ∫₀ᵗ eR(τ)dτ + ωL × (JL ωL) ``` **Inner Loop Carrier Trajectory Generation**: $$f_d(t) = G(R_L(t))^† w_d(t) + N(R_L(t))λ(t)$$ **Optimization Layer**: Dynamically adjusts internal force parameter $λ(t)$ to satisfy velocity constraints. ### Technical Innovations #### 1. Time-Varying Grasp Matrix Handling Unlike existing static load methods, this paper accounts for time-varying grasp matrix $G(t)$ and null space $N(t)$ caused by load attitude changes. #### 2. Mathematical Characterization of Carrier Velocity Constraints Derives explicit expression for carrier velocity: $$\dot{p}_{R_i}(t) = v_{L_i}(t) + \frac{L_i}{T_i(t)}(E_i(t) + G_i(t))$$ where $E_i(t)$ represents external contribution and $G_i(t)$ represents internal force contribution. The non-stopping constraint is equivalent to: $$0 \notin V_i := \{\dot{p}_{R_i}(t) : t \geq 0\}$$ #### 3. Online Optimization Algorithm Defines optimization problem: $$x^* = \arg\min_{x \in \mathbb{R}^2} J(x)$$ Subject to: $\|v_{L_i}(t) + \frac{L_i}{T_i}\Pi_i(t)(\dot{G}^†_i w_d + G^†_i \dot{w}_d + \dot{N}_i λ + N_i \dot{λ})\|^2 \geq \varepsilon$ The objective function includes smoothness and continuity penalty terms. ## Experimental Setup ### Experimental Platform 1. **Model Validation Experiments**: 3 Crazyflie 2.1 quadrotor unmanned aerial vehicles using Crazyswarm software and external motion capture system 2. **Numerical Simulation**: 4-carrier system considering complete load and cable dynamics, including cable elasticity and other non-ideal factors ### Load Specifications - **Experimental Load**: Triangular load constructed from carbon fiber tubes, 0.515m edge length, 0.012kg mass - **Cable Length**: Approximately 1m fine cotton line - **Simulated Load**: 1.0kg mass, inertia matrix 0.01I₃ kg·m² ### Trajectory Design Simulation employs composite trajectory: initial rest → linear motion along x-axis → final rest, specifically testing carrier behavior during load speed changes. ### Evaluation Metrics 1. **Load Tracking Accuracy**: Position error $e_p$ and attitude error $e_R$ 2. **Carrier Velocity Constraint**: $\|\dot{p}_{R_i}\| \geq \varepsilon = 0.2$ m/s 3. **Trajectory Smoothness**: Velocity profile continuity ## Experimental Results ### Model Validation Results Preliminary experiments (Figure 3) confirm: 1. **Trajectory Tracking Accuracy**: Multirotor unmanned aerial vehicles accurately track elliptical reference trajectories 2. **Load Stability**: Load attitude remains stable during carrier motion 3. **Model Validity**: Simplified theoretical model aligns closely with actual system behavior ### Simulation Comparison Results #### Without Optimization (Figures 5-7) - **Carrier Trajectory**: Sharp turns occur, causing instantaneous velocity to drop to zero - **Load Tracking**: Average position error 0.013m, good tracking performance - **Velocity Constraint**: Multiple violations of non-stopping constraint #### With Optimization (Figures 8-11) - **Carrier Trajectory**: Smooth elliptical trajectory avoiding sharp turns - **Load Tracking**: Average position error 0.015m, slightly higher but acceptable - **Velocity Constraint**: Strictly satisfies $\|\dot{p}_{R_i}\| \geq 0.2$ m/s - **Trajectory Characteristics**: Carriers execute loitering flight when load is stationary ### Key Findings 1. **Constraint Satisfaction**: Optimization algorithm successfully ensures carrier velocity remains above threshold 2. **Tracking Performance**: Velocity constraint introduction has minimal impact on load tracking accuracy 3. **Smoothness**: Optimized trajectories significantly improve smoothness of carrier motion ## Related Work ### Single Carrier Suspended Load Early research focused on single multirotor transporting suspended loads, primarily addressing swing suppression [5-7]. ### Cooperative Multirotor Systems Extended to multiple multirotor platforms for cooperative transportation, achieving six-degree-of-freedom load control [8-13], but dependent on hovering capability. ### Non-Stopping Carrier Theory Recent theoretical work [14,15] established feasibility of non-stopping carriers maintaining static load equilibrium, but limited to open-loop control. ### Positioning of This Work's Contribution This paper is the first to achieve closed-loop control for dynamic load tracking with non-stopping carriers, filling an important gap between theory and practical application. ## Conclusions and Discussion ### Main Conclusions 1. **Theoretical Breakthrough**: First achievement of closed-loop load trajectory tracking control with non-stopping flying carriers 2. **Technical Innovation**: Online optimization strategy effectively resolves coupling between carrier velocity constraints and load control 3. **Practical Validation**: Method performs well on complex trajectories (static and dynamic combinations) ### Limitations 1. **Simplified Carrier Dynamics**: Does not account for complete dynamic constraints of fixed-wing unmanned aerial vehicles 2. **Optimization Solvability**: Optimization problem may be infeasible in certain cases 3. **Experimental Scope**: Preliminary validation only on multirotor platforms ### Future Directions 1. **Refined Trajectory Planning**: Consider input saturation and carrier-specific state constraints 2. **Predictive Control**: Employ receding horizon methods replacing reactive optimization 3. **Real Fixed-Wing Validation**: Comprehensive experimental verification on actual non-stopping unmanned aerial vehicles 4. **Mechatronic Integration**: Address mechanical installation and adaptation of cables and loads ## In-Depth Evaluation ### Strengths 1. **Strong Innovation**: First solution to closed-loop load control with non-stopping carriers, possessing significant theoretical value 2. **Complete Methodology**: Forms complete technical chain from theoretical modeling through control design to experimental validation 3. **Mathematical Rigor**: Carrier velocity constraint characterization is rigorous, optimization problem design is reasonable 4. **Practical Value**: Provides feasible technical pathway for long-distance load transportation ### Weaknesses 1. **Limited Experimental Validation**: Verification only on multirotor platforms, lacking real fixed-wing unmanned aerial vehicle experiments 2. **Optimization Robustness**: Insufficient discussion of optimization problem feasibility and robustness 3. **Disturbance Handling**: Capability for handling external disturbances and model uncertainties requires further verification 4. **Computational Complexity**: Online optimization computational burden and real-time performance require more detailed analysis ### Impact 1. **Academic Contribution**: Initiates new research direction in non-stopping carrier cooperative control 2. **Application Prospects**: Provides important technical foundation for unmanned aerial vehicle logistics, rescue operations, and other applications 3. **Reproducibility**: Detailed method description and clear parameter settings ensure good reproducibility ### Applicable Scenarios 1. **Long-Distance Logistics**: Particularly suitable for load transportation tasks requiring long endurance 2. **Emergency Relief**: Material delivery to remote areas 3. **Marine Operations**: Material transportation between offshore platforms 4. **Military Applications**: Tactical supply delivery ## References This paper cites 16 relevant references covering suspended load transportation, multirotor cooperative control, fixed-wing unmanned aerial vehicle applications, and other key domains, providing solid theoretical foundation. Particularly important references include: - [14,15]: Pioneering work by Gabellieri and Franchi on non-stopping carrier static equilibrium - [8-13]: Classical research on multirotor cooperative transportation - [5-7]: Foundational theory of single carrier suspended load control --- **Overall Assessment**: This is a high-quality control theory paper making important contributions to the field of non-stopping flying carrier cooperative control. While experimental validation could be enhanced, its theoretical innovation and practical value make it a significant advance in the field.