Investigating Feed-Forward Back-Propagation Neural Network with Different Hyperparameters for Inverse Kinematics of a 2-DoF Robotic Manipulator: A Comparative Study

Year 2024, Volume: 6 Issue: 2, 90 - 110, 30.06.2024

Rania Bouzid , Hassène Gritli , Jyotindra Narayan

https://doi.org/10.51537/chaos.1375866

Cited By: 2

Abstract

Inverse kinematics is a significant challenge in robotic manipulators, and finding practical solutions plays a crucial role in achieving precise control. This paper presents a study on solving inverse kinematics problems using the Feed-Forward Back-Propagation Neural Network (FFBP-NN) and examines its performance with different hyperparameters. By utilizing the FFBP-NN, our primary objective is to ascertain the joint angles required to attain precise Cartesian coordinates for the end-effector of the manipulator. To accomplish this, we first formed three input-output datasets (a fixed-step-size dataset, a random-step-size dataset, and a sinusoidal-signal-based dataset) of joint positions and their respective Cartesian coordinates using direct geometrical formulations of a two-degree-of-freedom (2-DoF) manipulator. Thereafter, we train the FFBP-NN with the generated datasets using the MATLAB Neural Network Toolbox and investigate its potential by altering the hyperparameters (e.g., number of hidden neurons, number of hidden layers, and training optimizer). Three different training optimizers are considered, namely the Levenberg-Marquardt (LM) algorithm, the Bayesian Regularization (BR) algorithm, and the Scaled Conjugate Gradient (SCG) algorithm. The Mean Squared Error is used as the main performance metric to evaluate the training accuracy of the FFBP-NN. The comparative outcomes offer valuable insights into the capabilities of various network architectures in addressing inverse kinematics challenges. Therefore, this study explores the application of the FFBP-NNs in tackling the inverse kinematics, and facilitating the choice of the most appropriate network design by achieving a portfolio of various experimental results by considering and varying different hyperparameters of the FFBP-NN.

Keywords

Robotic manipulator, Inverse kinematics, Feed-Forward back propagation, Artifical neural network, Hyperparameters, Levenberg-Marquardt algorithm, Bayesian Regularization algorithm, Scaled Conjugate Gradient algorithm, Different Datasets, Mean squared error, R-value

References

• Abbas, M., J. Narayan, and S. K. Dwivedy, 2019 Simulation analysis for trajectory tracking control of 5-DOFs robotic arm using ANFIS approach. In 2019 5th International Conference On Computing, Communication, Control And Automation (ICCUBEA), pp. 1–6.
• Aravinddhakshan, S., S. Apte, and S. M. Akash, 2021 Neural network based inverse kinematic solution of a 5 DOF manipulator for industrial application. Journal of Physics: Conference Series 1969: 012010.
• Aysal, F. E., ˙I. Çelik, E. Cengiz, and Y. O˘guz, 2023 A comparison of multi-layer perceptron and inverse kinematic for RRR robotic arm. Politeknik Dergisi pp. 1–1.
• Becerra, G. and R. Kremer, 2011 Ambient intelligent environments and environmental decisions via agent-based systems. Journal of Ambient Intelligence and Humanized Computing 2: 185–200.
• Benavente-Peces, C., A. Ahrens, and J. Filipe, 2014 Advances in technologies and techniques for ambient intelligence.
• Bouzid, R., J. Narayan, and H. Gritli, 2023 Feedforward backpropagation artificial neural network for modeling the forward kinematics of a robotic manipulator. In 2023 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT), pp. 302–307, Sakheer, Bahrain.
• Bouzid, R., J. Narayan, and H. Gritli, 2024a Artificial neural networks for the forward kinematics of a SCARA manipulator: A comparative study with two datasets. In 2024 ASU International Conference in Emerging Technologies for Sustainability and Intelligent Systems (ICETSIS), pp. 1792–1797.
• Bouzid, R., J. Narayan, and H. Gritli, 2024b Exploring neural networks for forward kinematics of the robotic arm with different length configurations: A comparative analysis. In 2024 IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), volume 2, pp. 1–6.
• Bouzid, R., J. Narayan, and H. Gritli, 2024c Investigating neural network hyperparameter variations in robotic arm inverse kinematics for different arm lengths. In 2024 Third International Conference on Power, Control and Computing Technologies (ICPC2T), pp. 351–356.
• Bouzid, R., J. Narayan, and H. Gritli, 2024d Solving inverse kinematics problem for manipulator robots using artificial neural network with varied dataset formats. In Complex Systems and Their Applications, edited by E. Campos-Cantón, G. Huerta-Cuellar,
• E. Zambrano-Serrano, and E. Tlelo-Cuautle, pp. 55–78, Cham, Springer Nature Switzerland.
• Cagigas-Muñiz, D., 2023 Artificial neural networks for inverse kinematics problem in articulated robots. Engineering Applications of Artificial Intelligence 126: 107175.
• Cimen, M. E., Z. Garip, M. A. Pala, A. F. Boz, and A. Akgül, 2019 Modelling of a chaotic system motion in video with artificial neural networks. Chaos Theory and Applications 1: 38 – 50.
• Darba, A., N. B. Sushmi, and D. Subbulekshmi, 2022 Performance analysis of FFBP-LM-ANN based hourly GHI prediction using environmental variables: A case study in chennai. Mathematical Problems in Engineering 2022: 1713657.
• Dash, K. K., B. B. Choudury, and S. K. Senapati, 2017 A inverse kinematic solution of a 6-DOF industrial robot using ANN. Indian Journal of Scientific Research 15: 97–101.
• Del Rosario Martinez-Blanco, M., V. H. Castañeda-Miranda, G. Ornelas-Vargas, H. A. Guerrero-Osuna, L. O. Solis-Sanchez, et al., 2016 Generalized regression neural networks with application in neutron spectrometry. In Artificial Neural Networks, edited by J. L. G. Rosa, chapter 3, IntechOpen, Rijeka.
• Denavit, J. and R. S. Hartenberg, 1955 A kinematic notation for lower-pair mechanisms based on matrices. Journal of Applied Mechanics 22: 215–221.
• Di Pietro, A., D. Torresi, M. Zadro, L. Cosentino, C. Ducoin, et al., 2012 The inverse kinematics thick target scattering method as a tool to study cluster states in exotic nuclei. Journal of Physics: Conference Series 366: 012013.
• Duka, A.-V., 2014 Neural network based inverse kinematics solution for trajectory tracking of a robotic arm. Procedia Technology 12: 20–27.
• Dumitriu, D. N., O. D. Melinte, and M. Ionescu, 2020 Neural networks kinematics guidance of lewansoul learm 5r serial manipulator. Acta Electrotehnica 61.
• Ganapathy, S., 1984 Decomposition of transformation matrices for robot vision. Pattern Recognition Letters 2: 401–412.
• Gao, B., Z. Zhu, J. Zhao, and L. Jiang, 2017 Inverse kinematics and workspace analysis of a 3 DOF flexible parallel humanoid neck robot. Journal of Intelligent & Robotic Systems 87: 211–229.
• Gao, R., 2020 Inverse kinematics solution of robotics based on neural network algorithms. Journal of Ambient Intelligence and Humanized Computing 11: 6199–6209.
• García-Samartín, J. F. and A. Barrientos, 2023 Kinematic modelling of a 3RRR planar parallel robot using genetic algorithms and neural networks. Machines 11.
• Ghaleb, N. M. and A. A. Aly, 2018 Modeling and control of 2- DOF robot arm. International Journal of Emerging Engineering Research and Technology 6: 24–31.
• Handayani, A. N., N. Lathifah, H. W. Herwanto, R. A. Asmara, and K. Arai, 2018 Neural network bayesian regularization backpropagation to solve inverse kinematics on planar manipulator.
• In 2018 Joint 7th International Conference on Informatics, Electronics & Vision (ICIEV) and 2018 2nd International Conference on Imaging, Vision & Pattern Recognition (icIVPR), pp. 99–104, IEEE.
• Huo, L. and L. Baron, 2008 The joint-limits and singularity avoidance in robotic welding. Industrial Robot: An International Journal 35: 456–464.
• Ibarra-Pérez, T., J. M. Ortiz-Rodríguez, F. Olivera-Domingo, H. A. Guerrero-Osuna, H. Gamboa-Rosales, et al., 2022 A novel inverse kinematic solution of a six-DOF robot using neural networks based on the taguchi optimization technique. Applied Sciences 12: 9512.
• Jenhani, S., H. Gritli, and G. Carbone, 2022 Comparison between some nonlinear controllers for the position control of Lagrangiantype robotic systems. Chaos Theory and Applications 4: 179 – 196.
• Karaca, Y., 2023 Computational complexity-based fractional-order neural network models for the diagnostic treatments and predictive transdifferentiability of heterogeneous cancer cell propensity. Chaos Theory and Applications 5: 34 – 51.
• Kayri, M., 2016 Predictive abilities of Bayesian regularization and Levenberg–Marquardt algorithms in artificial neural networks: a comparative empirical study on social data. Mathematical and Computational Applications 21: 20.
• Kele¸s, Z., G. Sonugür, and M. Alçın, 2023 The modeling of the rucklidge chaotic system with artificial neural networks. Chaos Theory and Applications 5: 59 – 64.
• Kim, J. S., Y. H. Jeong, and J. H. Park, 2016 A geometric approach for forward kinematics analysis of a 3-sps/s redundant motion manipulator with an extra sensor using conformal geometric algebra. Meccanica 51: 2289–2304.
• Köker, R., C. Öz, T. Çakar, and H. Ekiz, 2004 A study of neural network based inverse kinematics solution for a three-joint robot. Robotics and autonomous systems 49: 227–234.
• Kumar, P. et al., 2018 Artificial neural network based geometric error correction model for enhancing positioning accuracy of a robotic sewing manipulator. Procedia Computer Science 133: 1048–1055.
• Lathifah, N., A. N. Handayani, H. W. Herwanto, and S. Sendari, 2018 Solving inverse kinematics trajectory tracking of planar manipulator using neural network. In 2018 International Conference on Information and Communications Technology (ICOIACT), pp. 483–488, IEEE.
• Li, H. and A. V. Savkin, 2018 An algorithm for safe navigation of mobile robots by a sensor network in dynamic cluttered industrial environments. Robotics and Computer-Integrated Manufacturing 54: 65–82.
• Liu,W., D. Chen, and J. Steil, 2017 Analytical inverse kinematics solver for anthropomorphic 7-DOF redundant manipulators with human-like configuration constraints. Journal of Intelligent & Robotic Systems 86: 63–79.
• Madhuraghava, P., B. D. Fakruddin, R. V. Subhash, and N. Sunil, 2018 Modelling and structural, analysis of a 6-DOF robot spray coating manipulator. The International Journal of Engineering and Science 7: 48–56.
• Mahajan, A., H. Singh, and N. Sukavanam, 2017 An unsupervised learning based neural network approach for a robotic manipulator. International Journal of Information Technology 9: 1–6.
• Martinez-garcia, J. A., A. M. Gonzalez-zapata, E. J. Rechy-ramirez, and E. Tlelo-cuautle, 2022 On the prediction of chaotic time series using neural networks. Chaos Theory and Applications 4: 94 – 103.
• Møller, M. F., 1993 A scaled conjugate gradient algorithm for fast supervised learning. Neural networks 6: 525–533.
• Narayan, J., M. Abbas, B. Patel, and S. K. Dwivedy, 2023 Adaptive RBF neural network-computed torque control for a pediatric gait exoskeleton system: an experimental study. Intelligent Service Robotics 232: 726–732.
• Narayan, J., S. Banerjee, D. Kamireddy, and S. K. Dwivedy, 2022 Fuzzy membership functions in ANFIS for kinematic modeling of 3R manipulator. In Handbook of Smart Materials, Technologies, and Devices: Applications of Industry 4.0, edited by C. M. Hussain and P. Di Sia, pp. 1101 – 119, Springer International Publishing, Cham.
• Narayan, J. and A. Singla, 2017a ANFIS based kinematic analysis of a 4-DOFs SCARA robot. In 2017 4th International Conference on Signal Processing, Computing and Control (ISPCC), pp. 205–211.
• Narayan, J. and A. G. Singla, 2017b Inverse Kinematic Study of Spatial Serial Manipulators using ANFIS Approach. Ph.D. thesis, Thapar Institute of Engineering and Technology.
• Narayan, J., E. Singla, S. Soni, and A. Singla, 2018 Adaptive neurofuzzy inference system–based path planning of 5-degrees-offreedom spatial manipulator for medical applications. Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine 232: 726–732.
• Noorani, I. and F. Mehrdoust, 2022 Parameter estimation of uncertain differential equation by implementing an optimized artificial neural network. Chaos, Solitons & Fractals 165: 112769.
• Petrescu, R. V., R. Aversa, B. Akash, R. Bucinell, J. Corchado, et al., 2017 Inverse kinematics at the anthropomorphic robots, by a trigonometric method. American Journal of Engineering and Applied Sciences 10: 394–411.
• Petrovi´c, L., 2018 Motion planning in high-dimensional spaces. arXiv preprint arXiv:1806.07457.
• Ranganathan, A., 2004 The levenberg-marquardt algorithm. Tutoral on LM algorithm 11: 101–110.
• Rea Minango, S. N. and J. C. E. Ferreira, 2017 Combining the stepnc standard and forward and inverse kinematics methods for generating manufacturing tool paths for serial and hybrid robots. International Journal of Computer Integrated Manufacturing 30: 1203–1223.
• Reiter, A., A. Müller, and H. Gattringer, 2018 On higher order inverse kinematics methods in time-optimal trajectory planning for kinematically redundant manipulators. IEEE Transactions on Industrial Informatics 14: 1681–1690.
• Snieder, R., 1998 The role of nonlinearity in inverse problems. Inverse problems 14: 387.
• Takatani, H., N. Araki, T. Sato, and Y. Konishi, 2019 Neural network-based construction of inverse kinematics model for serial redundant manipulators. Artificial Life and Robotics 24: 487–493.
• Theofanidis, M., S. I. Sayed, J. Cloud, J. Brady, and F. Makedon, 2018 Kinematic estimation with neural networks for robotic manipulators. In Artificial Neural Networks and Machine Learning– ICANN 2018: 27th International Conference on Artificial Neural Networks, Rhodes, Greece, October 4-7, 2018, Proceedings, Part III 27, pp. 795–802, Springer.
• Wagaa, N., H. Kallel, and N. Mellouli, 2023 Analytical and deep learning approaches for solving the inverse kinematic problem of a high degrees of freedom robotic arm. Engineering Applications of Artificial Intelligence 123: 106301.
• Wang, W., G. Yu, M. Xu, and D.Walker, 2014 Coordinate transformation of an industrial robot and its application in deterministic optical polishing. Optical Engineering 53: 055102–055102.
• Xu,W., Z. Mu, T. Liu, and B. Liang, 2017 A modified modal method for solving the mission-oriented inverse kinematics of hyperredundant space manipulators for on-orbit servicing. Acta Astronautica 139: 54–66.
• Zhao, D., Y. Bi, and Y. Ke, 2018 Kinematic modeling and inverse kinematics solution of a new six-axis machine tool for oval hole drilling in aircraft wing assembly. The International Journal of Advanced Manufacturing Technology 96: 2231–2243.