Application of a Narrow Neural Network Algorithm to Dynamic Analysis Results of a One-Link Planar Robotic Arm

Öz

The kinematics, dynamics, control, and optimal design of robotic mechanisms have been widely investigated using various analytical and computational approaches. In recent years, artificial neural network (ANN) algorithms have emerged as an additional tool in mechanism and robotics research. Experimental and theoretical results are commonly analyzed using supervised and unsupervised neural network algorithms. However, many existing studies do not explicitly examine the effects of input–output variations within the employed datasets. In this study, the dynamic modeling and simulation of a one-link planar robotic arm are first presented. Based on the simulation results, datasets are generated and analyzed using a neural network algorithm. One of the primary objectives of this study is to demonstrate the importance of appropriate input–output (predictor–response) selection in neural network applications for robotic mechanisms. A further objective is to contribute ongoing scientific investigations about robotic modeling which use neural network algorithms. The results show that different choices of input and output variables can lead to significantly different prediction performances, highlighting the critical role of dataset formulation in neural network–based robotic analysis.

Anahtar Kelimeler

• ANN
• dynamic analysis
• narrow neural network
• one-link mechanism
• robotic arm

Dar Yapılı Bir Yapay Sinir Ağı Algoritmasının Tek Bağlantılı Düzlemsel Bir Robot Kolunun Dinamik Analiz Sonuçlarına Uygulanması

Öz

Robotik mekanizmaların kinematiği, dinamiği, kontrolü ve optimal tasarımı çeşitli analitik ve hesaplamalı yaklaşımlar kullanılarak geniş ölçüde incelenmiştir. Son yıllarda, yapay sinir ağı (YSA) algoritmaları mekanizma ve robotik araştırmalarında ek bir araç olarak ortaya çıkmıştır. Deneysel ve teorik sonuçlar yaygın olarak denetimli ve denetimsiz sinir ağı algoritmaları kullanılarak analiz edilmektedir. Ancak, mevcut çalışmaların birçoğu kullanılan veri kümeleri içerisindeki girdi–çıktı değişimlerinin etkilerini açıkça incelememektedir. Bu çalışmada, tek bağlantılı (one-link) düzlemsel bir robot kolunun dinamik modellemesi ve simülasyonu öncelikle sunulmaktadır. Simülasyon sonuçlarına dayanarak veri kümeleri oluşturulmuş ve bir sinir ağı algoritması kullanılarak analiz edilmiştir. Bu çalışmanın temel amaçlarından biri, robotik mekanizmalarda sinir ağı uygulamalarında uygun girdi–çıktı (tahmin edici–yanıt) seçiminin önemini göstermektir. Bir diğer amaç ise sinir ağı algoritmalarını kullanan robotik modelleme konusundaki devam eden bilimsel çalışmalara katkıda bulunmaktır. Sonuçlar, farklı girdi ve çıktı değişkeni seçimlerinin önemli ölçüde farklı tahmin performanslarına yol açabileceğini göstermekte ve sinir ağı tabanlı robotik analizde veri kümesi oluşturmanın kritik rolünü vurgulamaktadır.

Anahtar Kelimeler

• YSA
• dinamik analiz
• Dar Yapılı Yapay Sinir Ağı
• Tek Bağlantılı mekanizma
• robotik kol

Kaynakça

1. [1] Spong, M. W., Hutchinson, S., & Vidyasagar, M. (2020). Robot modeling and control (2nd ed.). Wiley.
2. [2] Söylemez, E. (2018). Makina teorisi 2: Makine dinamiği. Birsen Yayınevi.
3. [3] Waldron, K. J., Kinzel, G. L., & Agrawal, S. K. (2016). Kinematics, dynamics, and design of machinery (3rd ed.). John Wiley & Sons.
4. [4] Abduljabbar, Z., ElMadany, M. M., & Al-Dokhiel, H. D. (1993). Controller design of a one-link flexible robot arm. Computers & Structures, 49(1), 117–126. https://doi.org/10.1016/0045-7949(93)90130-6
5. [5] Raju, E. M., Krishna, L. S., Mouli, Y. S., & Rao, V. N. (2015). Effect of link flexibility on tip position of a single link robotic arm. Journal of Physics: Conference Series, 662(1), 012020. https://doi.org/10.1088/1742-6596/662/1/012020
6. [6] Yang, G. B., & Donath, M. (1988, April). Dynamic model of a one-link robot manipulator with both structural and joint flexibility. In Proceedings of the IEEE International Conference on Robotics and Automation (pp. 476–481). IEEE. https://doi.org/10.1109/ROBOT.1988.12097
7. [7] De Luca, A., & Siciliano, B. (1989). Trajectory control of a nonlinear one-link flexible arm. International Journal of Control, 50(2), 699–715. https://doi.org/10.1080/00207178908953460
8. [8] Endo, T., Matsuno, F., & Kawasaki, H. (2014). Force control and exponential stabilisation of one-link flexible arm. International Journal of Control, 87(9), 794–807. https://doi.org/10.3182/20050703-6-CZ-1902.01313

9. [9] Shchuka, A., & Goldenberg, A. A. (1989). Tip control of a single-link flexible arm using a feedforward technique. Mechanism and Machine Theory, 24(5), 439–455. https://doi.org/10.1016/0094-114X(89)90072-4
10. [10] M.E. Demir, and R.K. Ergün, "Mechanical and wear performance of Al-, mica-, SiO2-filled glass fiber-reinforced composites and prediction of wear properties with artificial neural networks," In Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering, p. 09544089241307840, 2 Jan. 2025, doi: 10.1177/095440892413078
11. [11] M.E. Demir, and R.K. Ergün, "Mechanical and wear performance of Al-, mica-, SiO2-filled glass fiber-reinforced composites and prediction of wear properties with artificial neural networks," In Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering, p. 09544089241307840, 2 Jan. 2025, doi: 10.1177/095440892413078.
12. [12] Demir, M. E., Çetkin, E., Ergün, R. K., & Denizhan, O. (2024). Tribological and mechanical properties of nanofilled glass fiber reinforced composites and analyzing the tribological behavior using artificial neural networks. Polymer Composites, 45(5), 4233–4249. https://doi.org/10.1002/pc.28055
13. [13] Demir, M. E. (2025). The effect of filler type (tungsten carbide, zinc oxide) and content on the mechanical and wear behavior of jute/flax reinforced epoxy hybrid composites: Experimental and artificial neural network analysis. Polymer Composites. https://doi.org/10.1002/pc.29879
14. [14] Gürbüz, H., Demir, M. E., Baday, Ş., & Akcan, İ. H. (2025). Yapay sinir ağı ile mikro parçacık dolgulu kompozitlerin tribolojik özelliklerinin tahmini. Journal of Materials and Mechatronics: A, 6(1), 68–82. https://doi.org/10.55546/jmm.1581662
15. [15] Denizhan, O. (2025). Application of various artificial neural network algorithms for regression analysis in the dynamic modeling of a three-link planar RPR robotic arm. Machines, 13(11), 1031. https://doi.org/10.3390/machines13111031
16. [16] Denizhan, O. (2024). Comparison of different supervised learning algorithms for position analysis of the slider-crank mechanism. Alexandria Engineering Journal, 92, 39–49. https://doi.org/10.1016/j.aej.2024.02.055
17. [17] Kong, F. G., Li, Q., & Zhang, W. J. (1999). An artificial neural network approach to mechanism kinematic chain isomorphism identification. Mechanism and Machine Theory, 34(2), 271–283. https://doi.org/10.1016/S0094-114X(98)00035-4
18. [18] Efe, M. O., & Kaynak, O. (2000). Stabilizing and robustifying the learning mechanisms of artificial neural networks in control engineering applications. International Journal of Intelligent Systems, 15(5), 365–388. https://doi.org/10.1002/(SICI)1098-111X(200005)15:5
19. [19] Yang, Z., Wu, J., & Mei, J. (2007). Motor-mechanism dynamic model based neural network optimized computed torque control of a high speed parallel manipulator. Mechatronics, 17(7), 381–390. https://doi.org/10.1016/j.mechatronics.2007.04.009
20. [20] Cui, H. (2025). High-dimensional learning of narrow neural networks. Journal of Statistical Mechanics: Theory and Experiment, 2025(2), 023402. https://doi.org/10.1088/1742-5468/adb1d6
21. [21] Beise, H. P., Da Cruz, S. D., & Schröder, U. (2021). On decision regions of narrow deep neural networks. Neural Networks, 140, 121–129. https://doi.org/10.1016/j.neunet.2021.02.024
22. [22] Lu, L., Su, Y., & Karniadakis, G. E. (2018). Collapse of deep and narrow neural nets. arXiv preprint arXiv:1808.04947. https://doi.org/10.48550/arXiv.1808.04947
23. [23] Babu, M. V., & Banana, K. (2024). A study on narrow artificial intelligence—An overview. International Journal of Engineering Science and Advanced Technology, 24(4), 210–219
24. [24] Alali, M., Sharef, N. M., Murad, M. A., Hamdan, H., & Husin, N. A. (2019). Narrow convolutional neural network for Arabic dialects polarity classification. IEEE Access, 7, 96272–96283. https://doi.org/10.1109/ACCESS.2019.2929208
25. [25] Lee, H., Kim, Y., Yang, S. Y., & Choi, H. (2024). Improved weight initialization for deep and narrow feedforward neural network. Neural Networks, 176, 106362. https://doi.org/10.1016/j.neunet.2024.106362
26. [26] Dommel, J., & Wegner, S. A. (2025). An in-depth look at approximation via deep and narrow neural networks. Neurocomputing. https://doi.org/10.1016/j.neucom.2025.131769