Optimizing ANN Hyperparameters with Metaheuristic Algorithms for Inverse Kinematics of a 3-DoF Rehabilitation Exoskeleton Robot

Rania Bouzid; Hasséne Gritli; Jyotindra Narayan

doi:10.2991/978-94-6463-654-3_9

<Previous Article In Volume

Next Article In Volume>

Optimizing ANN Hyperparameters with Metaheuristic Algorithms for Inverse Kinematics of a 3-DoF Rehabilitation Exoskeleton Robot

Authors

Rania Bouzid¹^{, 2}, Hasséne Gritli²^{, 3}^{, *}, Jyotindra Narayan⁴^{, 5}

¹Polytechnic School of Tunisia, University of Carthage, B.P. 743, 2078, La Marsa, Tunisia

²Laboratory of Robotics, Informatics and Complex Systems (RISC Lab - LR16ES07), National Engineering School of Tunis, University of Tunis, El Manar, BP. 37, Le Belvédère, 1002, Tunis, Tunisia

³Higher Institute of Information and Communication Technologies, University of Carthage, Technopole of Borj Cédria, Route de Soliman, BP 123, Hammam Chatt 1164, Ben Arous, Tunisia

⁴Department of Mechanical Engineering, Indian Institute of Technology Patna, Patna, 801106, India

⁵Department of Computing, Imperial College, London, SWS 2AZ, UK

^*Corresponding author. Email: grhass@yahoo.fr Email: hassene.gritli@istic.ucar.tn

Corresponding Author

Hasséne Gritli

Available Online 24 February 2025.

DOI: 10.2991/978-94-6463-654-3_9 How to use a DOI?
Keywords: Artificial Neural Network (ANN); Inverse kinematics; Rehabilitation exoskeleton robot; Random dataset; Sinusoidal dataset; Hyperparameters; Metaheuristic algorithms
Abstract: This paper presents a novel approach for optimizing the hyperparameters of Artificial Neural Networks (ANNs) using metaheuristic algorithms to solve the inverse kinematics problem for a 3-DoF rehabilitation exoskeleton robot. The exoskeleton aims to assist in rehabilitation by enabling accurate control of upper limb movement, which is critical for patients recovering motor functions. Two distinct datasets are employed: a random step size dataset and a sinusoidal signal dataset, reflecting different movement patterns in rehabilitation scenarios. The focus is on finding optimal values of three critical hyperparameters of the ANN: activation function, hidden layer size, and learning rate. Metaheuristic algorithms, specifically Particle Swarm Optimization (PSO) and Genetic Algorithm (GA), are utilized to determine the optimal configuration of these hyperparameters. The performance of the proposed method is evaluated by comparing the Mean Squared Error (MSE) across both datasets. Our results demonstrate the effectiveness of the proposed approach in enhancing the precision and reliability for the exoskeleton control, with the lowest MSE values achieved being 0.083801 for the GA-ANN algorithm and 0.040729 for the PSO-ANN algorithm on the sinusoidal signal-based dataset.
Open Access: Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

<Previous Article In Volume

Next Article In Volume>

Volume Title: Proceedings of the International Conference on Decision Aid and Artificial Intelligence (ICODAI 2024)
Series: Atlantis Highlights in Intelligent Systems
Publication Date: 24 February 2025
ISBN: 978-94-6463-654-3
ISSN: 2589-4919
DOI: 10.2991/978-94-6463-654-3_9 How to use a DOI?
Open Access: Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

ris enw bib

TY  - CONF
AU  - Rania Bouzid
AU  - Hasséne Gritli
AU  - Jyotindra Narayan
PY  - 2025
DA  - 2025/02/24
TI  - Optimizing ANN Hyperparameters with Metaheuristic Algorithms for Inverse Kinematics of a 3-DoF Rehabilitation Exoskeleton Robot
BT  - Proceedings of the  International Conference on Decision Aid and Artificial Intelligence (ICODAI 2024)
PB  - Atlantis Press
SP  - 103
EP  - 120
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6463-654-3_9
DO  - 10.2991/978-94-6463-654-3_9
ID  - Bouzid2025
ER  -

download .riscopy to clipboard