Skip to Content
AerospaceAerospace
  • Article
  • Open Access

14 July 2026

28 Pages

A Research Method for Dynamic Coupling Modeling and Analysis of Space Robots with Multi-Flexibility Cooperative Structural Deformation

and
1
Shaanxi Key Laboratory of Advanced Manufacturing and Health Management for Aviation Components, Shaanxi University of Technology, Hanzhong 732000, China
2
College of Intelligent Manufacturing, Long Dong University, Qingyang 745000, China
*
Author to whom correspondence should be addressed.
This article belongs to the Section Astronautics & Space Science

Abstract

Space robots must be designed with a large operating radius and lightweight components (flexible joints and flexible links). The integration of flexible components inevitably affects the control accuracy of the robot. When modeling the link deformation of the space robot, the idealized description method has the problem of incomplete accuracy in link deformation due to the coupling effects of the non-fixed base, flexible joints, and changes in control parameters. To address this challenge and improve the accuracy of robot modeling, this study proposes a dynamic coupling analysis method for the link deformation of space robots based on time-domain response. The double finite difference method is adopted to calculate the time-domain response of link deformation. On this basis, using the local linear operator method, the modal information of the deformed links is extracted. This method systematically quantifies the individual and interactive effects of the robot’s base motion, joint flexibility, and control parameters on link deformation. This work not only lays a theoretical foundation for the accurate dynamic modeling of space robots but also provides practical insights for optimizing their structural design. Ultimately, it contributes to the development of more reliable and high-performance space robot systems.

1. Introduction

Space robots can assist astronauts in accomplishing tasks such as spacecraft docking, maintenance, and capture [1,2]. As spacecraft develop toward the trends of large size, light weight, and high precision, space robots are required to feature low stiffness, large working radius, and integrated actuation. These requirements lead to the deformation of the robot links. The joint structure incorporates harmonic reducers, elastic dampers, and flexible sensors, which endow the joints with flexibility. Consequently, space robots are classified as a type of multi-flexible special-purpose robot [3]. The flexible components significantly complicate the modeling and control of space robots [4,5]. The description of the flexibility of these components will greatly affect the modeling accuracy. Therefore, to achieve better control of flexible space robots, a more accurate modeling and analysis model is essential.
To establish an accurate dynamic model for a flexible-link robot, it is first essential to describe and analyze the deformation of its links in a discrete manner. For links with regular geometric shapes, the Euler–Bernoulli beam is commonly employed as a standard beam model. The Assumed Modes Method (AMM) is typically used to describe such deformations; however, it neglects the shear effect [6]. In contrast, the finite element method (FEM) is highly effective in characterizing the deformation of irregularly shaped links—and additionally, it can account for the shear effect [7]. Beyond the two aforementioned methods for describing the deformation of flexible links, there exist other approaches, including the Modal Synthesis Method (MSM), Lumped Parameter Method (LPM), Finite Segment Method (FSM), and B-spline Interpolation (BSI) [8,9]. When these methods are applied to flexible robot modeling, their boundary conditions still need to be evaluated; nevertheless, over-idealization of boundary conditions may lead to deviations between the model and the actual system [10]. Therefore, the modal parameters of the system must be determined based on the deformation response of the flexible links under actual operating conditions. Hao et al. [11] presented an assumed mode analysis-based manipulator modeling method adopting fixed-load boundary conditions for links. The ideal constraints differ greatly from practical conditions, causing low modal extraction accuracy. Zhou et al. [12] detected manipulator link deformation through experiments, yet the method demands massive repeated tests and ignores effects of joints and base platform, leading to obvious deviations. Kumar et al. [13] established a mobile manipulator dynamic model via the extended Hamilton principle with joint–rotor inertia coupling, but neglected the mobile platform’s impact on arm deformation. Summarizing the literature [10,11,12,13], it is evident that under idealized assumptions, the deformation of flexible robotic arms during multi-modal information acquisition deviates significantly from actual descriptions. This paper treats space robots as mobile flexible robotic arms, comprehensively considering the effects of the base, flexible joints, and geometric shape on the deformation of flexible linkages. This approach ensures boundary conditions better align with real-world operational scenarios, enabling more accurate extraction of modal information from the time-domain response of link deformation. This approach not only provides a theoretical foundation for dynamic modeling of flexible space robots but also enables analysis of link deformation patterns under various influencing factors.
As summarized above, the process of constructing a dynamic model for flexible robots involves two key steps: First, identifying, evaluating, and extracting modal information related to the deformation of flexible links is fundamental to ensuring the model accurately reflects flexible characteristics. Second, combining the physical parameters and structural features of the manipulator is necessary to further extract additional data required for predicting the robot’s dynamic behavior. Based on this process, various modeling techniques have been developed and applied to the dynamics of flexible link robots. Examples include the Newton–Euler method using the Modal Superposition Method (MSM), the Lagrangian method employing the Absolute Node Coordinate Method (AMM), and the Lagrangian method utilizing the finite element method (FEM) [14,15]. Kraïem et al. [16] used the AMM and Newton–Euler methods to construct a single-joint flexible robot dynamics model. While this model can rapidly capture changes in the robot’s motion state, it exhibits significant shortcomings: the end-effector positioning error of the flexible manipulator remains relatively high, primarily due to insufficient description accuracy of the flexible link. Masoumi et al. [17] established a dynamic model for flexible robots based on the Automatic Market Maker (AMM) and Lagrange methods. While this model incorporates load effects as an advantage, it fails to account for joint structural damping and load interactions within the deformation boundary conditions of its flexible linkages. Consequently, the model’s description of flexible components deviates from actual operating conditions. Shang et al. [18] proposed a robotic dynamics model applicable to complex-shaped link structures and employed finite element dynamics analysis to describe link deformation. This approach suffers from two limitations: first, an irreconcilable trade-off exists between computational efficiency and analytical accuracy; second, its applicability is restricted, rendering it unsuitable for modeling scenarios involving multi-link flexible robots. Ouyang et al. [19] established a 2DOF flexible robot for modeling research, incorporating matching boundary conditions for link bending deformation into the model. However, this model does not account for interactions between multiple links in the consideration of link deformation, presenting certain limitations. Liu et al. [20] employed the Absolute Node Coordinate Method to construct a nonlinear model of the flexible mechanical system (SLFM). However, their study did not fully account for the coupling effects between flexible deformation and rigid motion, a factor that may impact the accuracy of the dynamic analysis. Summarizing and analyzing the references [16,17,18,19,20], based on current research focuses, existing studies on dynamic modeling of flexible robotic arms predominantly concentrate on deformation description and modeling method optimization for single flexible links (e.g., References [16,17]). However, research on modeling multi-flexible-link systems, particularly in boundary condition design and coordinated deformation description under multi-link coupling scenarios, remains significantly underdeveloped (e.g., References [18,19]). This paper comprehensively considers multiple factors—mobile platforms, joint flexibility, and adjacent joint characteristics—to modify the deformation boundaries of flexible linkages. This approach enables more accurate descriptions of the deformation in multi-flexible robotic arms, thereby enhancing the precision of dynamic modeling for flexible robotic systems.
Based on the above research, dynamic modeling of flexible robots requires the incorporation of flexible link deformation with high-fidelity characterization, which is a crucial prerequisite for guaranteeing modeling accuracy. Deformation acquisition for flexible manipulators mainly relies on theoretical modeling and experimental testing, yet both techniques suffer from inherent limitations. The analytical modal method (AMM) adopts idealized hypotheses and continuous deformation functions to characterize structural deformation, which inevitably produces deviations from the practical deformation of flexible links under real working scenarios. Alternatively, the finite element method (EFM) demands linearization and fine discretization of flexible structures, leading to intensive computational costs. Moreover, conventional simulation tools fail to fully capture the complex dynamic behaviors of spatial flexible robots, further compromising the modeling reliability of simulation-based approaches. In terms of experimental investigation, the extreme complexity of space environments hinders the replication of in-orbit operating conditions, restricting the direct applicability of ground-based experimental conclusions to space flexible robot systems. In addition, high-precision dynamic measurement instruments can only acquire one-dimensional deformation information, which substantially limits the comprehensiveness and effectiveness of experimental testing. To address the aforementioned deficiencies, this work considers the multi-factor coupling effects on flexible link deformation and develops a novel deformation description method for spatial flexible robot components using the time-domain response data of flexible manipulators.
In this paper, a dynamic coupling analysis model of robot link deformation is constructed with a flexible space robot as the research object. The paper is organized as follows: Part one, introduction, Part two, problem description; Part three, complex dynamics of link deformation of flexible space robot; and analysis of link deformation based on control characteristics; Part four, identification of link modes based on dynamic time-domain response; Part five, numerical simulation validation of the model; and Part six, analytical summary of this paper.

2. Problem Description

Recapitulating earlier studies on flexible robotics, it is evident that, in addition to its extensive range of rigid body motions, the modeling of a general flexible robot needs account for the deformation of the flexible pieces attached to it [21]. Regarding the issue of describing the deformation of flexible links, the majority of research that has been done thus far has used cantilever beams as standard beams, whose boundaries consider the fixed–freedom combination boundary as the ideal boundary condition [22,23]. The AMM is used to describe the deformation of these links. This is not at all like the description of link deformation found in actual space robots that operate [24].
  • The space robot base and the manipulator are coupled to each other, and their motion affects the deformation of the manipulator link.
  • A flexible drive mechanism is designed into the joint structure of the manipulator, which causes significant torsional deformation of the joint and a coupling effect between it and the manipulator [25].
  • When a space robot performs trajectory tracking control, the drive torque will cause its actual tracking trajectory to be continuously adjusted in a closed loop around the predetermined trajectory, which will lead to the inevitable influence of the control parameters on the deformation of the manipulator link [26].
Therefore, to establish a more accurate dynamic model of the flexible space robot, this paper comprehensively considers factors such as robot base, joint flexibility damping and control parameters, and analyzes its link boundaries and deformation characteristics.

5. Simulation and Experiment

5.1. Model Verification and Analysis

To verify the validity of the article mode, a 2DOF flexible space robot in Figure 3 on a mobile platform is taken as the research object [13]; it is a three-dimensional space model. Firstly, o0-x0y0z0 is the inertial coordinate, and oB-xByBzB is the center of the base coordinate. The first joint of the manipulator is installed at the oByB axis of the base and is offset by 0.75 m. The coordinate is oi-xiyizi (i = 1,2), and the joint central axis coincides with oizi. The joint structure is equipped with an elastic damping structure whose equivalent stiffness is ki and the equivalent damping is ci. The manipulator link is fixed on the wheel hub, and the axis oixi coincides with the axis of the link when the link without deformation. An actuator with a two-finger structure at the end of the manipulator arm has a concentrated mass of 10 kg. The gravity coefficient in this paper is 9.8 × 10−4 N/kg. The overall structure of the space robot is shown in Figure 5, and the specific parameters are shown in Table 1.
Figure 5. The overall structure of 2DOF flexible space robot.
Table 1. Physical parameters of flexible space robot.
In Table 1, the mechanical arms B1 and B2 are aerospace-grade carbon fiber material. The shear modulus within the core plane of this material is typically 3.8–5.6 GPa. In this article the shear modulus is 7.5 GPa [28].
In MATAB R2023b, the joint initial values are θ1 = θ2 = 0. The initial deformation value of the link is ui = vi = wi = γji = 0, (j = x, y, z). The torque is r τ 1 e = − 15 sin 3 t N·m and τ 2 e = 15 sin 0.3 t N·m, respectively. By the actual convergence results of the simulation, the simulation step size is adjusted, and it is Δt = 2.5 × 10−2 s. The simulation process is calculated by odel5s, and the simulation time is 10 s.
To verify the effectiveness of the link deformation analysis method, the dynamic Coupled Analysis Model (DCAM), the flexible joint–flexible link analysis model under ideal hypothesis mode (FJ-FL-AMM) [29] and the analysis model of flexible joint–rigid manipulator (FJ-RM) are simulated and compared [30]. The correctness and calculation accuracy of the classical modeling method (FJ-RM) are verified, and can be used as a reference model for the verification of the modeling method in this paper. Firstly, in the FJ-FL-AMM system, the link deformation is described by AMM [31]. The link boundary condition at one end is fixed and the other end is free, while the initial conditions are 0. Then, the characteristics of the flexible joint are described by the lumped parameters. The dynamics model of the 2DOF flexible space robot is constructed based on the Lagrange method. Secondly, in the FJ-RM system, the manipulator link is regarded as a rigid link, and the dynamics model of the flexible joint–rigid link robot is established by using the Newton–Euler method. Finally, three system models are simulated by MATAB R2023b.
By the curve changes in Figure 6, there is no significant change in the posture of the pedestal calculated by the three models within 0 ≤ t ≤ 3 s. After 3 s, compared with the FJ-RM modeling method, the posture of the robot base in DCAM changes greatly. The maximum relative error is 8.55%. The error is lower than the 10% allowable threshold commonly used in the aerospace field [12]. But their general trends are basically the same. This verifies the correctness of the model presented in this paper. Meanwhile, compared with the FJ-FL-AM, the pose of the base in DCAM changes greatly, which proves that the flexible joint also has a certain influence on the base’s stable operation. Especially after 5 s, the curve deviation under the two models gradually becomes larger, which proves that there is an interaction between the base movement and the flexible manipulator, which is also consistent with the previous research conclusions [29] (coupling analysis of the flexible base and the manipulator). This further proves the correctness of DCAM.
Figure 6. Pose changes of space robot base under different analysis models: Position: (a) X-direction, (b) Y-direction, (c) Z-direction; attitude: (d) X-axis, (e) Y-axis, (f) Z-axis.
By the analysis of the manipulator joints in Figure 7, the angle change trend of the manipulator joint under the three different analysis models is consistent. Under the action of sinusoidal driving torque, the link deformation counteracts the manipulator operation, which makes the peak value of the joint angle fluctuate. Compared with the joint response of FJ-FL-AMM, the joint response in DCAM is more obvious. The curve changes in Figure 6 show the changes in robot joint flexibility under different FJ-FL coupling structures. When the FJ-RM model comprehensively considers joint flexibility and damping, and the ratio between equivalent stiffness and equivalent damping is constant, the joint flexibility changes periodically. When the deformable elastic force of the link reacts on the joint structure, the joint flexibility changes will be intensified, especially under DCAM. Theoretically, the link deformation in DCAM will increase the joint angle change, which conforms to its energy conservation law. And it will also act on the flexible joint change, which also conforms to its energy conservation.
Figure 7. Joint change for space robots under different analysis models: (a) joint angle 1, (b) joint angle 2, (c) joint flexibility 1, (d) joint flexibility 2.
By the flexible variation curves of link 1 and link 2 in Figure 8, it can be seen that the study analyzed the manipulator link deformation from a three-dimensional perspective. By the setting of the floating coordinate and the joint coordinate, compared with the axial tensile deformation in the X-direction in Figure 8a,d and the shear deformation in the Z-direction in Figure 8c,f, the transverse bending deformation in the Y-direction in Figure 8b,e is larger, with an order-of-magnitude difference of 103, which is in accordance with the bending deformation of links with Euler–Bernoulli beams as the standard beams. The above analysis proves the correctness of the dynamic coupling analysis model of link deformation for flexible space robots established in this paper.
Figure 8. Flexible vibration of space robot link under different analysis models: Link 1: (a) X direction, (b) Y direction, (c) Z direction; Link 2: (d) X direction, (e) Y direction, (f) Z direction.
During the operation of the actual space robot, the coupling link deformation is larger when considering the robot base motion, joint flexibility, damping and other factors, which is consistent with the principle of the system energy conservation. Although the system control input and the system action are relatively smooth, with the change in time t, the coupling link bending deformation will appear suddenly (3.5 s, 5.5 s and 7.5 s in Figure 8b,e). This is due to the rapid change in the coupling link mass geometry leading to the system’s instability. As the changes in the quadratic terms of the joints trajectories for the robot system are large, they can obtain large accelerations (2 s, 4 s, 6 s, and 8 s in Figure 6a,b). And as the curvature of the robot base trajectories changes drastically (6 s and 8 s in Figure 4), the link deformations of the flexible space robots produce large changes. This further proves the necessity of the dynamic coupling analysis of link deformation, and indirectly proves the correctness of its analytical model.
This study is not only applicable to the flexible deformation analysis of large components such as space robots and solar sail panels, but also to the flexible deformation analysis of operators on mobile platforms such as interplanetary exploration vehicles, large boom cranes and large marine manipulators. It integrates the non-fixed carrier base motion, complex joint structure and large-size body deformation for analysis and research, and is significantly superior to the single body deformation analysis.

5.2. Numerical Simulation and Characteristic Analysis

To analyze the link deformation characteristics of flexible manipulator under the control law, this section takes the robot shown in Figure 3 as the object for calculation and analysis. The joint’s starting value is θ1 = θ2 = 0. The initial value of the link deformation is u i = v i = w i = γ j i = 0 , (j = x, y, z). The displacement and torsion angle of the robot link deformation can be obtained either directly or indirectly during the robot’s actual operation. The feedback control algorithm is used to control the stable operation of the robot base, and the control parameters are kept unchanged in the subsequent simulation. Finally, different control parameters are set by Equation (42). Using the solution method in Section 3.3, the link deformation is compared. The pose control parameters of the base are KBP= diag([KBp,KBv]), KBp= diag([25,25,25]) and KBv= diag([2,2,2]). When the link deformation of the manipulator is not suppressed, three different combination parameters are used to design the joint control parameters in Table 2. To further analyze the change law of the influence of control parameters on the link deformation, the control parameter pair is taken as the horizontal coordinate variable, while the system frequency and deformation response under the control law are taken as the longitudinal variable, and their values are further simulated, as shown in Figure 9.
Table 2. Changes in link deformation in frequency domain and time domain under different control parameters.
Figure 9. The output variables of the mechanical system under different control parameters: (a) Link 1, (b) Link 2.
When determining the tip load mass of the link, the data in column 3 is the base frequency of the cantilever beam under fixed-end load boundary conditions. Based on this, the vibration frequency of the first link and the second link in column 3 can be quantified. And the vibration foundation frequency of the link decreases with the increase in the link tip mass. As the tip mass of the link is 4 times higher in two different cases, its base frequency is reduced by 50%. This is consistent with the research results in the literature [27] (the study gives the research values and results of the flexible robot arm at 0 kg and 3 kg), and the results of the above research are consistent with the classical vibration theory. The data in column 5 are the vibration frequencies of the EBTT coupled system under the elastic damped end-load boundary. This system increases with the increase in joint stiffness, but its end transverse bending v(li) decreases [32]. The joint damping of the robot is 0.05, and the joint stiffness is the value in column 4 in Table 2. Based on the above analysis, compared with the research results in reference [33] (which provides stiffness parameter design and robust adjustment parameters of flexible manipulator), the research and analysis results in this paper are more comprehensive and specific.
Combining the constrained boundary conditions and joint stiffness coefficients in Equation (43), the vibration frequency and the link tip bending displacement in the control system can be calculated by using the solution methods in Section 3.3 and Section 3. The calculated results are the values in columns 8 to 10 in Table 2, and are compared and analyzed with the values of the cantilever beam and coupling system in Table 2. When the system control parameters (Kp and Kd) are small, the link vibration frequency in the control system is smaller compared to the link vibration frequency in the cantilever beam or coupled system. And link vibration frequency is reduced by about 40%. According to the analysis of the curve change trend in Figure 8, the flexible link deformation decreases with the increase in the control parameters. And the link vibration frequency will gradually approach the foundation frequency of the cantilever beam in the control system. This is because within the range of allowable control parameters, KP increase accelerates the instantaneous response speed of the system, and Kd decrease reduces the overshoot parameters, and this overcomes the system oscillation of the manipulator. By the above analysis, the link vibration frequency in the control system is lower than that of the cantilever beam and the coupling system, although the joint structure has a damping constraint effect with the increase in control parameters. The manipulator link vibration tends to be similar to that of the cantilever beam with lower frequency because the control parameters cannot be increased infinitely, thus improving the stability of the whole robot system.

5.3. Numerical Simulation and Modal Analysis

To verify the validity of the modal identification model, the second link of the manipulator in Figure 3 is used as the object. And the time-domain response of the manipulator link deformation is calculated by using the solution method in Section 2. Twelve test points are selected in the axial direction of the link, and the deformation response acceleration of each monitoring point is obtained. Then, under the pulse excitation signal with a sampling frequency of 100 Hz and sampling time of 10 s, the numerical simulation is carried out on the selected monitoring points. The natural frequency, modal shape and damping ratio of the flexible link are calculated by using the time-domain response mode identification method in Section 4. Meanwhile, ANSYS Workbench 2024 R2 is used to simulate the modal analysis of the link deformation. Firstly, the three-dimensional model of the space robot is constructed by Solidworks and it is imported into Ansys. In the modal analysis module, different boundary conditions are set at the two ends of the robot link. The boundary condition at one end is a rotatable spring damping structure, and the other end boundary condition is a concentrated mass load. Secondly, in the axial direction of the manipulator, the deformation and displacement of the manipulator at 12 monitoring points are selected for monitoring. The response data of the link deformation can be obtained by the monitoring data. On this basis, the modal analysis module can be used to extract the vibration frequency, modal damping ratio and modal mode of the link deformation. Finally, the modal identification results of finite element analysis are compared with those of this paper, and the specific results are shown in Table 3. Modal parameters under different recognition methods are compared in Figure 10. On this basis, using the modal confidence function method in Ref [6], the correlation of the modal information identified by the above two methods is analyzed. The validity of the modal information identified in this paper is evaluated.
Table 3. Comparison of modal parameters based on different methods.
Figure 10. Modal parameters of link deformation under different identification methods: (a) Frequency, (b) Damping ratio, (c) Mode displacement.
Comparative analysis of the frequencies under this paper’s modal identification method and the finite element method through Table 3 and Figure 10a reveals that the maximum relative error between the two is not more than 8.55%, which verifies that the modal identification of the frequencies in this paper is feasible. By comparing the damping ratios of the two recognition methods in Figure 10b, it can be seen that the maximum relative difference of the damping ratios of the two modes is 4%, which verifies the correctness of the damping ratios of the mode recognition. Figure 10c is used to compare the modal displacement of the flexible manipulator end with the two identification methods, and it is found that the maximum displacement difference between the two methods is 5%, which further verifies the correctness of the mode identification method proposed in this paper. Based on the confidence criterion of modal recognition, it can be seen that the correlation confidence between the modal recognition method in this paper and the finite element method exceeds 0.93 and approaches 1. Therefore, we can consider that the modal frequencies and modal modes obtained by the two methods are basically the same, which proves the validity of the modal information recognition method in this paper. Based on the modal information recognition method presented in this paper, the modal parameters of the link deformation response for flexible space robots in complex systems can be identified. By using the identified modal information, the link deformation of flexible space robots is described, which is very beneficial to accurate robot dynamics modeling.
Based on the above Section 5.1, Section 5.2 and Section 5.3, the advantages of the model established in this paper are further clarified.
Compared with the traditional FJ-FL-AMM and FJ-RM modeling methods, the proposed DCAM rigid–flexible coupling analysis method has multiple verified advantages reflected by all groups of simulation curves and tabular data: as shown in Figure 6 base pose curves, this method revises link boundary conditions to fully incorporate floating base inertia, joint stiffness and damping, accurately capturing the mutual rigid–flexible excitation between the spacecraft platform and manipulator that traditional ideal cantilever-based models fail to reproduce.
Figure 8 link vibration results prove it adopts complete 3D Timoshenko beam theory and double finite difference parallel solving to comprehensively calculate axial stretch, bidirectional bending and torsional deformation of links, overcoming the 2D deformation limitation of the Assumed Modes Method; Table 2 and Figure 9 control parameter simulations reveal it innovatively embeds PD closed-loop feedback torque into dynamic constraint equations, enabling quantitative analysis of how proportional and differential gains affect system natural frequency and terminal vibration amplitude, a capability absent from open-loop traditional models.
Figure 7 joint flexibility curves show it can fully reflect the coupling amplification effect of link elastic counter-torque on joint torsional oscillation in line with energy conservation rules, unlike rigid-link FJ-RM which ignores all arm vibration; Table 3 modal comparison data verifies its LLE time-domain modal identification algorithm generates frequency, damping and mode shape results with a maximum relative error of only 8.55% against high-precision Ansys FEM and MAC coefficients above 0.93, delivering more real structural modal information than artificially preset ideal modes; meanwhile, the multi-core parallel double finite difference calculation flow in Figure 3 balances computational accuracy and efficiency better than fine-mesh finite element methods, so overall this method realizes more authentic physical characterization, higher modal identification precision and quantifiable multi-factor coupling analysis unavailable to traditional simplified models, perfectly matching the actual on-orbit operating conditions of space robots.

6. Conclusions

In this paper, the deformation description of a space flexible manipulator under dynamic coupling is studied.
Firstly, the flexible space robot is equivalent to a flexible manipulator model of a movable platform. On this basis, joint flexibility, damping and load effects are introduced into the boundary conditions of the link, and the complex boundary conditions are modified. Based on the principle of energy conservation and moment balance, a dynamic coupling analysis model of the link deformation for space robot with coupling structure FJ-FL is proposed. By the double finite difference method, the solution flow of the link deformation is given and the time-domain response is calculated. Secondly, in the complex boundary conditions of link deformation, the closed-loop control parameters are introduced, and the dynamic constraint equation of link deformation is established and solved. Then, the time-domain response of link deformation is used as a dataset, and the modal recognition method of LLAE is used to identify the modal information of link deformation of a space manipulator. Finally, numerical simulation is used to verify the correctness of the analytical model. Compared with the classical link analysis models FJ-FL-AMM and FJ-RM, the dynamic coupling analysis model proposed in this chapter comprehensively considers the influences of robot base, joint flexibility, joint damping and other factors on link deformation, so as to obtain a response closer to the actual working condition. With an increase in control parameters, the flexible deformation of the link decreases. In the control system, the vibration frequency of the link will gradually approach the base frequency of the cantilever beam. Based on the confidence criterion of modal recognition, it can be seen that the correlation confidence between the modal recognition method in this paper and the finite element method exceeds 0.93 and approaches 1. On this basis, the modal information recognition method based on time-domain response in this paper has a strong practical significance for establishing the dynamic model of flexible space robots. And this lays a theoretical foundation for the accurate dynamic modeling of subsequent space robots.

Author Contributions

F.Z.: Conceptualization, Supervision, Data curation, Investigation, Formal analysis, Writing—original draft, Funding acquisition, Resources. N.L.: Writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

Shaanxi University of Technology talent launched a special project (Funder: Shaanxi University of Technology; Funding Number: SLGRC202417), Project of the University Engineering Research Center of the Shaanxi Provincial Department of Education (Funder: Shaanxi Provincial Department of Education; Funding Number: 24JR044), Key Laboratory Project of Advanced Manufacturing and Health Management for Aviation Components in Shaanxi Province (Funder: Advanced Manufacturing and Health Management Laboratory for Aviation Components in Shaanxi Province; Funding Number: SLGKFKT03).

Data Availability Statement

All data generated or analyzed during this study are included in this published article and they can be both public and effective.

Conflicts of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

  1. Xi, H.; Chen, B.; Tian, J. Dual-observer-based composite robust vibration control for a rigid-flexible hybrid space robot-target combination. Nonlinear Dyn. 2025, 113, 16545–16569. [Google Scholar] [CrossRef] [Scilit]
  2. Dong, Z.; Wang, L.; Zhu, Y. Motion planning of free-flying space robots for tracking relative motion tumbling targets by sequential optimization via reinforcement learning. Acta Astronaut. 2025, 235, 463–474. [Google Scholar] [CrossRef] [Scilit]
  3. Yang, Y.; Su, Q.; Hui, W. Adaptive model-free finite-time tracking control for flexible-link manipulator with parametric and non-parametric uncertainties. Nonlinear Dyn. 2025, 113, 13339–13355. [Google Scholar] [CrossRef] [Scilit]
  4. Yang, Y.; Zhang, Z.X.; Liu, M.; Han, H.H.; Wei, J.; Cao, D.Q. Dynamic modeling and analysis for a large flexible spacecraft with antenna trusses and solar wings. Appl. Math. Model. 2026, 150, 116405. [Google Scholar] [CrossRef] [Scilit]
  5. Shang, D.; Fan, X.; Yin, M. Dynamic modeling and speed control for flexible manipulator with elastic joint based on fractional-order disturbance observer. J. Braz. Soc. Mech. Sci. Eng. 2025, 47, 513. [Google Scholar] [CrossRef] [Scilit]
  6. Ji, N.; Huang, Z.Y.; Liu, J.K. Bilateral consistency and vibration control for nonlinear flexible Timoshenko robotic arms in signed directed communication graph with quantized control inputs. Commun. Nonlinear Sci. Numer. Simul. 2026, 152, 109334. [Google Scholar] [CrossRef] [Scilit]
  7. Yu, J.W.; Wu, M.Y.; Ji, J.C.; Yang, W.H. Neural Network-Based Region Tracking Control for a Flexible-Joint Robot Manipulator. J. Comput. Nonlinear Dyn. 2024, 19, 021003. [Google Scholar]
  8. Zhang, Y.; He, L.; Chen, J. Neural Network Adaptive Hierarchical Sliding Mode Control for the Trajectory Tracking of a Tendon-Driven Manipulator. Chin. J. Mech. Eng. 2025, 38, 18. [Google Scholar] [CrossRef] [Scilit]
  9. Gupta, N.; Pratiher, B. Hydrodynamic modeling with cascade PID control and experimental realization of a two-link flexible underwater robotic manipulator. Ocean Eng. 2025, 335, 121545. [Google Scholar] [CrossRef] [Scilit]
  10. Yan, X.; Shi, L.; Shan, M. Vibration-suppressed trajectory planning for flexible space manipulators during on-orbit operations. Nonlinear Dyn. 2025, 113, 20977–21000. [Google Scholar] [CrossRef] [Scilit]
  11. Hao, C.F.; Dong, L.L.; Yan, J. Modal recognition method combining LLE and HHT based on test response. Proc. Chin. Mech. Soc. 2017, 12, 12–36. [Google Scholar]
  12. Zhou, Y.P.; Lou, J.Q.; Chen, T.H. Identification and experiment of coupling dynamics model of flexible arm system driven by servo joint. J. Vib. Shock. 2019, 38, 278–284. [Google Scholar]
  13. Pravesh, K.; Barun, P. Modal characterization with nonlinear behaviors of a two-link flexible manipulator. Arch. App. Mech. 2019, 89, 1201–1220. [Google Scholar] [CrossRef] [Scilit]
  14. Do, T.T.; Vu, V.H. Linearization of dynamic equations for vibration and modal analysis of flexible joint manipulators. Mech. Mach. Theory 2022, 167, 104516. [Google Scholar] [CrossRef] [Scilit]
  15. Li, C.Y.; Wang, X.; Cheng, Y.; O’Regan, D. Boundary stabilization for a nonlinear flexible beam under an adaptive fractional-order proportional-integral control. Commun. Nonlinear Sci. Numer. Simul. 2026, 152, 109352. [Google Scholar] [CrossRef] [Scilit]
  16. Kraïem, S.; Rognant, M.; Waitman, S.; Chelikh, I.; Biannic, J.M. Dynamics and robust control of a space manipulator system servicing a satellite with flexible appendages. IFAC J. Syst. Control 2025, 33, 100328. [Google Scholar] [CrossRef] [Scilit]
  17. Masoumi, M.; Kohansal, N.; Eftekhari, M. Nonlinear vibration of flexible two links manipulator with variable cross sections. Commun. Nonlinear Sci. Numer. Simul. 2026, 152, 109250. [Google Scholar] [CrossRef] [Scilit]
  18. Shang, D.Y.; Liu, J.Q.; Yin, M.; Li, X.P. Speed fluctuation suppression strategy for flexible joint-flexible link robotic manipulators via servo motor measurement feedback signal. Measurement 2026, 257, 118871. [Google Scholar] [CrossRef] [Scilit]
  19. Ouyang, Q.H.; Chen, B.; Xia, H.Y.; Liu, H.X.; Tian, J.; Zhang, Y.Z. Optimal trajectory planning for coupled vibration suppression of a flexible base and space flexible manipulator. Aerosp. Sci. Technol. 2025, 165, 110511. [Google Scholar] [CrossRef] [Scilit]
  20. Liu, Y.; Yao, X.Q.; Zhao, W. Distributed neural-based fault-tolerant control of multiple flexible manipulators with input saturations. Automatica 2023, 156, 111202. [Google Scholar] [CrossRef] [Scilit]
  21. Su, J.; Liu, L.X. Simulation research on active control of low frequency modal vibration of double-joint flexible manipulator. Int. J. Ind. Syst. Eng. 2021, 39, 287–298. [Google Scholar] [CrossRef] [Scilit]
  22. Yang, X.; Shen, L.; Jia, P.H. A Flexible FEM-BEM-DDM for EM Scattering by Multiscale Anisotropic Objects. IEEE Trans. Antennas Propag. 2021, 69, 8562–8573. [Google Scholar] [CrossRef] [Scilit]
  23. Wang, J.C.; Liu, J.K.; Jia, B.; He, Y.D.; Xia, S.G.; Zhou, Y.P. Vibration suppression and boundary control for nonlinear flexible rotating manipulator in three-dimensional space subject to output restrictions. Commun. Nonlinear Sci. Numer. Simul. 2023, 120, 107151. [Google Scholar] [CrossRef] [Scilit]
  24. Ma, X.Y.; Liu, J.K. Event-triggered boundary consensus force control for PDE modeling multi-flexible manipulators with both actuator delay and communication delay. ISA Trans. 2025, 160, 72–86. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  25. Zhang, X.Y.; Tang, L.; Liu, Y.J. Adaptive constraint control for flexible manipulator systems modeled by partial differential equations with dead-zone input. Int. J. Adapt. Control Signal Process. 2021, 35, 1404–1416. [Google Scholar] [CrossRef] [Scilit]
  26. Yao, W.; Guo, Y.; Wu, Y.F.; Guo, J. Robust Adaptive Dynamic Surface Control of Multi-link Flexible Joint Manipulator with Input Saturation. Int. J. Control Autom. Syst. 2022, 20, 577–588. [Google Scholar] [CrossRef] [Scilit]
  27. Ji, N.; Liu, J.K. Consensus of flexible Timoshenko manipulators with both disturbance observers and event-triggered distributed estimators. ISA Trans. 2025, 167, 65–77. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  28. Xu, X.Y.; Peng, G.Q. Material performance, manufacturing methods, and engineering applications in aviation of carbon fiber reinforced polymers: A comprehensive review. Thin-Walled Struct. 2025, 209, 112899. [Google Scholar] [CrossRef] [Scilit]
  29. Tavakkol, S.; Son, S.Y.; Lynett, P. Adaptive third order Adams-Bashforth time integration for extended Boussinesq equations. Comput. Phys. Commun. 2021, 265, 108006. [Google Scholar] [CrossRef] [Scilit]
  30. Mamo, M.B. Control of flexible link manipulators using nonlinear feedback. Results Eng. 2025, 27, 106186. [Google Scholar] [CrossRef] [Scilit]
  31. Du, Y.F. Research on Dynamic Modeling and Vibration Control of Flexible Space Robot. Harbin Inst. Technol. 2020, 2, 23–60. [Google Scholar]
  32. Gupta, N.; Pratiher, B. Dynamic modeling and effective vibration reduction of dual-link flexible manipulators with two-stage cascade PID and active torque actuation. Mech. Mach. Theory 2025, 205, 105867. [Google Scholar] [CrossRef] [Scilit]
  33. Ni, S.; Chen, W.; Wei, Z. Optimal trajectory planning of space flexible-link manipulator based on singular perturbation theory. J. Sound Vib. 2025, 617, 119242. [Google Scholar] [CrossRef] [Scilit]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Multiple requests from the same IP address are counted as one view.