Dynamics Modeling of Robot Manipulators Based on Deep Lagrangian Network and Torque Separation Technique

Abstract

In recent years, learning robot manipulator dynamics with deep networks has been extensively studied, as it avoids deriving the analytical expression of the robot dynamics equations. In particular, deep Lagrangian networks that incorporate the prior knowledge of Lagrangian mechanics into the deep networks have shown prominent advantages in robot dynamics modeling. Inspired by this, this paper proposes a robot dynamics modeling approach based on a deep Lagrangian network and a torque separation technique. First, friction modeling is incorporated into deep Lagrangian networks to more comprehensively learn the dynamics of the robot manipulator. Meanwhile, the inputs of deep Lagrangian networks are reconstructed using the prior knowledge of robot joint types, which can entail valuable information for the model learning process that may result in better approximation and generalization. Moreover, a torque separation technique is introduced to extract the inertial–Coriolis–centrifugal torques and gravity–friction torques from the input torques, which allows us to accurately estimate each part of the robot dynamics and helps us understand the in-depth dynamic characteristics of robot manipulators. The simulation results on two-degrees-of-freedom (2-DOF) and 5-DOF robot manipulators demonstrate the feasibility and effectiveness of the proposed dynamics modeling method.

Keywords

deep Lagrangian network / dynamics modeling / robot manipulator / torque separation technique

References

[1] S. Hu, Y. Wan, and X. Liang, “Adaptive Nonsingular Fast Terminal Sliding Mode Trajectory Tracking Control for Robotic Manipulators With Model Feedforward Compensation,” Nonlinear Dynamics 113, no. 13 (2025): 16893–16911.

[2] X. Liang, Z. Yao, W. Deng, et al., “Adaptive Control of n-Link Hydraulic Manipulators With Gravity and Friction Identification,” Nonlinear Dynamics 111, no. 20 (2023): 19093–19109.

[3] A. Chemori, R. Kouki, and F. Bouani, “A New Fast Nonlinear Model Predictive Control of Parallel Manipulators: Design and Experiments,” Control Engineering Practice 130 (2023): 105367.

[4] D. Lim, D. Kim, and J. Park, “Momentum Observer-Based Collision Detection Using LSTM for Model Uncertainty Learning,” in IEEE International Conference on Robotics and Automation (ICRA) (2021), 4516–4522.

[5] d. C. C. Wit, B. Siciliano, and G. Bastin, Theory of Robot Control (Springer Science & Business Media, 2012).

[6] Y. Huang, J. Ke, X. Zhang, et al., “Dynamic Parameter Identification of Serial Robots Using a Hybrid Approach,” IEEE Transactions on Robotics 39, no. 2 (2022): 1607–1621.

[7] P. Cao, Y. Gan, and X. Dai, “Model-Based Sensorless Robot Collision Detection Under Model Uncertainties With a Fast Dynamics Identification,” International Journal of Advanced Robotic Systems 16, no. 3 (2019): 1729881419853713.

[8] R. Ortega, V. Nikiforov, and D. Gerasimov, “On Modified Parameter Estimators for Identification and Adaptive Control. A Unified Framework and Some New Schemes,” Annual Reviews in Control 50 (2020): 278–293.

[9] S. Wu, Z. Li, W. Chen, et al., “Dynamic Modeling of Robotic Manipulator via an Augmented Deep Lagrangian Network,” Tsinghua Science and Technology 29, no. 5 (2024): 1604–1614.

[10] A. S. Polydoros, L. Nalpantidis, and V. Krüger, “Real-Time Deep Learning of Robotic Manipulator Inverse Dynamics,” in IEEE International Conference on Robotics and Automation (ICRA) (2015), 3442–3448.

[11] N. Yilmaz, J. Y. Wu, P. Kazanzides, et al., “Neural Network Based Inverse Dynamics Identification and External Force Estimation on the da Vinci Research Kit,” in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (2020), 1387–1393.

[12] D. Kim, D. Lim, and J. Park, “Transferable Collision Detection Learning for Collaborative Manipulator Using Versatile Modularized Neural Network,” IEEE Transactions on Robotics 38, no. 4 (2021): 2426–2445.

[13] M. Moradi and H. Malekizade, “Neural Network Identification Based Multivariable Feedback Linearization Robust Control for a Two-Link Manipulator,” Journal of Intelligent & Robotic Systems 72, no. 2 (2013): 167–178.

[14] H. Su, W. Qi, C. Yang, et al., “Deep Neural Network Approach in Robot Tool Dynamics Identification for Bilateral Teleoperation,” IEEE Robotics and Automation Letters 5, no. 2 (2020): 2943–2949.

[15] Z. Zhang and Z. Chen, “Modeling and Control of Robotic Manipulators Based on Symbolic Regression,” IEEE Transactions on Neural Networks and Learning Systems 34, no. 5 (2021): 2440–2450.

[16] X. Liang, Z. Yao, Y. Ge, et al., “Disturbance Observer Based Actor–Critic Learning Control for Uncertain Nonlinear Systems,” Chinese Journal of Aeronautics 36, no. 11 (2023): 271–280.

[17] M. Lutter, C. Ritter, and J. Peters, “Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learning,” arXiv preprint arXiv:1907.04490 (2019).

[18] M. Lutter and J. Peters, “Combining Physics and Deep Learning to Learn Continuous-Time Dynamics Models,” International Journal of Robotics Research 42, no. 3 (2023): 83–107.

[19] J. K. Gupta, K. Menda, Z. Manchester, et al., “A General Framework for Structured Learning of Mechanical Systems,” arXiv preprint arXiv:1902.08705 (2019).

[20] M. R. Kermani, M. Wong, R. V. Patel, et al., “Friction Compensation in Low and High-Reversal-Velocity Manipulators,” in IEEE International Conference on Robotics and Automation (ICRA) (2004), 4320–4325.

[21] Y. Choi, D. Jung, J. Cheong, et al., “Dynamics Superposition Method for Separating Torque Components of Robot Manipulators Using Sinusoidal Trajectories,” Journal of Mechanical Science and Technology 31 (2017): 5505–5513.

[22] D. Jung, H. Do, T. Choi, et al., “Robust Parameter Estimation of Robot Manipulators Using Torque Separation Technique,” IEEE Access 9 (2021): 150443–150458.

[23] F. L. Lewis, D. M. Dawson, and C. T. Abdallah, Robot Manipulator Control: Theory and Practice (CRC Press, 2003).

[24] J. Swevers, W. Verdonck, and J. De Schutter, “Dynamic Model Identification for Industrial Robots,” IEEE Control Systems Magazine 27, no. 5 (2007): 58–71.

[25] Z. Lin, “Riemannian Geometry of Symmetric Positive Definite Matrices via Cholesky Decomposition,” SIAM Journal on Matrix Analysis and Applications 40, no. 4 (2019): 1353–1370.

[26] A. Dalla Libera and R. Carli, “A Data-Efficient Geometrically Inspired Polynomial Kernel for Robot Inverse Dynamic,” IEEE Robotics and Automation Letters 5, no. 1 (2019): 24–31.

[27] J. Swevers, C. Ganseman, D. B. Tukel, et al., “Optimal Robot Excitation and Identification,” IEEE Transactions on Robotics and Automation 13, no. 5 (1997): 730–740.

[28] X. Yu, Y. Li, S. Zhang, et al., “Estimation of Human Impedance and Motion Intention for Constrained Human–Robot Interaction,” Neurocomputing 390 (2020): 268–279.

[29] D. Nguyen-Tuong, M. Seeger, and J. Peters, “Computed Torque Control With Nonparametric Regression Models,” in American Control Conference (2008), 212–217.

[30] Z. Yao, X. Liang, S. Wang, et al., “Model-Data Hybrid Driven Control of Hydraulic Euler–Lagrange Systems,” IEEE/ASME Transactions on Mechatronics 30, no. 1 (2025): 131–143.

[31] Z. Du, X. Chen, Q. Zhang, et al., “An Extended State Observer-Based Sliding Mode Control Method for Hydraulic Servo System of Marine Stabilized Platforms,” Ocean Engineering 279 (2023): 114386.

[32] X. Liang, Z. Yao, W. Deng, et al., “Adaptive Neural Network Finite-Time Tracking Control for Uncertain Hydraulic Manipulators,” IEEE/ASME Transactions on Mechatronics 30, no. 1 (2025): 645–656.

[33] D. P. Kingma and J. Ba, “Adam: A Method for Stochastic Optimization,” arXiv preprint arXiv:1412.6980 (2014).

[34] G. Boschetti and T. Sinico, “Designing Digital Twins of Robots Using Simscape Multibody,” Robotics 13, no. 4 (2024): 62.

[35] D. Lim, M. J. Kim, J. Cha, et al., “MOB-Net: Limb-Modularized Uncertainty Torque Learning of Humanoids for Sensorless External Torque Estimation,” International Journal of Robotics Research 44, no. 1 (2025): 96–128.

RIGHTS & PERMISSIONS

2025 The Author(s). International Journal of Mechanical System Dynamics published by John Wiley & Sons Australia, Ltd on behalf of Nanjing University of Science and Technology.