Implementation of ANN on Microcontrollers in Line Follower Robots
DOI:
https://doi.org/10.33795/jartel.v16i1.9544Keywords:
Arduino, Machine Learning, Line follower robot, Control System, Artificial Neural Network (ANN)Abstract
Line follower robots commonly use conventional control methods such as if–else logic and PID control, which suffer from limited flexibility, poor adaptability to changing track conditions, and require complex parameter tuning. This study proposes a machine learning–based control system using an Artificial Neural Network (ANN) with a feedforward architecture and two hidden layers. The ANN model is trained using supervised learning with sensor data and motor output obtained from initial simulations, then implemented on an Arduino microcontroller for real-time control. Experimental results under both bright and completely dark lighting conditions show that the ANN-based system can control the robot stably and responsively, achieving a 90% success rate in completing the test track. These results demonstrate that ANN improves navigation accuracy, motion consistency, and adaptability, while also being effectively deployable on microcontrollers with limited computational resources.
References
S. Saadatmand, S. Azizi, M. Kavousi, and D. C. Wunsch, “Autonomous Control of a Line Follower Robot Using a Q-Learning Controller,” 2020.
S. K. Risandriya and J. A. Tarigan, “Optimasi Sensor pada Robot Line Follower dengan Jenis Lapangan Berbeda dengan Metode Neural Network.”
M. Education, “Microcontroller Based Line Follower Robot 1,” vol. 11, no. 3, pp. 2730–2735, 2020.
O. Erbay, “Line Follower Robot with PID Control,” no. January, 2024.
H. M. Leal, R. S. Barbosa, and I. S. Jesus, “Control of a Mobile Line-Following Robot Using Neural Networks,” pp. 1–26, 2025.
C. Minaya, R. Rosero, M. Zambrano, and P. Catota, “Application of Multilayer Neural Networks for Controlling a Line-Following Robot in Robotic Competitions,” vol. 18, 2024, doi: 10.14313/JAMRIS/1.
A. Latif, H. A. Widodo, R. Rahim, and K. Kunal, “Implementation of Line Follower Robot Based Microcontroller ATMega32A,” pp. 15–20, 2020, doi: 10.18196/jrc.1316.
R. Beresford, “Basic Concepts of Artificial Neural Network (ANN) Modeling and Its Application in Pharmaceutical Research,” vol. 22, pp. 717–727, 2000.
T. M. Mitchell, Machine Learning.
Trivusi, “Algoritma Feedforward Neural Network: Pengertian dan Cara Kerjanya.” [Online]. Available: https://www.trivusi.web.id/2022/07/algoritma-feedforward-neural-network.html#
W. Pav and J. Campa, “Intelligent Control System for Line Follower Robot: From Theory to Implementation on the Mechatronics Field,” 2024, doi: 10.1109/DASC/PiCom/CBDCom/Cy59711.2023.10361359.
R. A. Sarhan and M. S. Hassan, “Arduino-Based Implementation of Kinematics for a 4 DOF Robot Manipulator Using Artificial Neural Network,” 2024, doi: 10.29354/diag/184235.
A. Salim and W. S. Pambudi, “Implementasi Metode Hybrid Artificial Neural Network (ANN)–PID untuk Perbaikan Proses Berjalan pada Prototype Robot Material Handling,” vol. 1, no. 3, pp. 155–164, 2015.
I. Pramana and D. Futra, “Implementasi Algoritma Q-Learning pada Robot Line Follower,” pp. 1–6, 2021.
A. E. Muñoz-Zavala, J. E. Macías-Díaz, and D. Alba-Cuéllar, “A Literature Review on Some Trends in Artificial Neural Networks for Modeling and Simulation with Time Series,” pp. 1–45, 2024.
