Absolute Magnetic Encoder Design Based On RBF Neural Networks
Keywords:
Rotary encoder, magnetic encoder, multisensor data fusion, hall effect sensor, ANN, RBF neural networks, analog angular measurement.Abstract
This paper proposes absolute magnetic encoder design for analog angular measurement using multi-sensor data-fusion based on Radial Basis Function (RBF) neural networks. Multiple linear Hall effect sensors and a magnet are used to realize the analog angular output. RBF neural networks are used to approximate multi-dimensional nonlinear function between the sensor values and angular position of the magnet. The parameters of the RBF network are determined by supplying the data for multiple sensor values and the corresponding angular position of the magnet. Trained RBF neural network can be used to obtain the analog angle output for the given sensor inputs and it can be implemented using 8 or 16-bit microcontroller. This design of the encoder allows flexibility in terms of placement of the sensors.
References
- Lee, B. Lee, K. Park, and R. Elmasri, “Fusion Techniques for Reliable Information: A Survey,” International Journal of Digital Content Technology and its Applications, vol. 4, no. 2, pp. 74–88, 2010.
- T. Le, H. V. Hoang, and J. W. Jeon, “Efficient method for correction and interpolation signal of magnetic encoders,” in Industrial Informatics, 2008. 6th IEEE International Conference on, 2008, pp. 1383–1388.
- Arthur Noronha Montanari and Everthon de Souza Oliveira, "A Novel Analog Multisensor Design Based on Fuzzy Logic: A Magnetic Encoder Application", IEEE Sensors Journal, DOI 10.1109/JSEN.2017.2752959.
- -S. Jang and Chuen-Tsai Sun, “Neuro-fuzzy modeling and control,” Proceedings of the IEEE, vol. 83, no. 3, pp. 378–406, 1995.
- https://en.wikipedia.org/wiki/Artificial_neural_network
- https://en.wikipedia.org/wiki/Radial_basis_function_network
Downloads
Published
Issue
Section
License
Copyright (c) IJSRCSEIT
This work is licensed under a Creative Commons Attribution 4.0 International License.