Design of a Tracking Controller Based on Machine Learning

Abstract

Tracking control of multibody systems is a challenging task requiring detailed modeling and control expertise. Especially in the case of closed-loop mechanisms, inverse kinematics as part of the controller may become a game stopper due to the extensive calculations required for solving nonlinear equations and inverting complicated functions. The procedure introduced in this paper substitutes such advanced human expertise by artificial intelligence through the utilization of surrogates, which may be trained from data obtained by classical simulation. The necessary steps are demonstrated along a parallel mechanism called λ-robot. Based on its mechanical model, the workspace is investigated, which is required to set proper initial conditions for generating data covering the used operation space of the robot. Based on these data, artificial neural networks are trained as surrogates for inverse kinematics and inverse dynamics. They provide forward control information such that the remaining error behavior is governed by a linear ordinary differential equation, which allows applying a linear quadratic regulator (LQR) from linear control theory. An additional feedback loop of the tracking error accounts for model uncertainties. Simulation results validate the applicability of the proposed concept.

Keywords

artificial neural network / inverse dynamics / inverse kinematics / machine learning / tracking control

References

[1] W. Schiehlen and P. Eberhard, Applied Dynamics (Springer, 2014).

[2] X. Rui, G. Wang, and J. Zhang, Transfer Matrix Method for Multibody Systems: Theory and Applications (Wiley, 2018).

[3] Society of Automotive Engineers, SAE J3016: Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles (SAE, 2014).

[4] K. Ogata, Modern Control Engineering (Prentice Hall, 2020).

[5] F. Ansarieshlaghi and P. Eberhard, “Disturbance Compensator for a Very Flexible Parallel Lambda Robot in Trajectory Tracking,” in 17th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2020) (2020), 394–401, https://doi.org/10.5220/0009790203940401.

[6] A. Isidori, Nonlinear Control Systems (Springer, 1995).

[7] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning (MIT Press, 2016).

[8] A. Hashemi, G. Orzechowski, A. Mikkola, and J. McPhee, “Multibody Dynamics and Control Using Machine Learning,” Multibody System Dynamics 58 (2023): 397–431.

[9] M. H. Korayem and S. R. Nekoo, “State-Dependent Differential Riccati Equation to Track Control of Time-Varying Systems With State and Control Nonlinearities,” ISA Transactions 57 (2015): 117–135.

[10] M. Pikuliński, P. Malczyk, and R. Aarts, “Data-Driven Inverse Dynamics Modeling Using Neural-Networks and Regression-Based Techniques,” Multibody System Dynamics, ahead of print, September 6, 2024, https://doi.org/10.1007/s11044-024-10024-2.

[11] H. Ren and P. Ben-Tzvi, “Learning Inverse Kinematics and Dynamics of a Robotic Manipulator Using Generative Adversarial Networks,” Robotics and Autonomous Systems 124 (2020): 103386.

[12] A. S. Polydoros, L. Nalpantidis, and V. Krüger, “Real-Time Deep Learning of Robotic Manipulator Inverse Dynamics,” in International Conference on Intelligent Robots and Systems (IROS 2015) (2015), 3442–3448, https://doi.org/10.1109/IROS.2015.7353857.

[13] M. Burkhardt, R. Seifried, and P. Eberhard, “Experimental Studies of Control Concepts for a Parallel Manipulator With Flexible Links,” Journal of Mechanical Science and Technology 29 (2015): 2685–2691.

[14] T. Kurz, P. Eberhard, C. Henninger, and W. Schiehlen, “From Neweul to Neweul-M2: Symbolical Equations of Motion for Multibody System Analysis and Synthesis,” Multibody System Dynamics 24 (2010): 25–41.

[15] “Matlab—Math. Graphics. Programming,” Mathworks, accessed November 24, 2024, https://www.mathworks.com/products/matlab.html.

[16] P. Geurts, D. Ernst, and L. Wehenkel, “Extremely Randomized Trees,” Machine Learning 63 (2006): 3–42.

[17] D. Bestle and T. Bielitz, “Real-Time Models for Systems With Costly or Unknown Dynamics,” PAMM 24, no. 2 (2024): 1–8, https://doi.org/10.1002/pamm.202400008.

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.