EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Proposing Two Neural Networks Structures for Enhancing Laser Sensor Distance Accuracy

Karam Samir Khalid, Rand Zuhair Khaleel, Hind Zuhair Khaleel, Amjad Jaleel Humaidi

Tikrit Journal of Engineering Sciences · 15.09.2026 · 33 (2) · 1-11

DOI: 10.25130/tjes.33.2.5 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Tikrit Journal of Engineering Sciences, online erschienen am 15.09.2026. Verfasst von Karam Samir Khalid und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The precision of distance measurement using a laser sensor significantly influences robot motion accuracy, localization, and obstacle avoidance. The present study proposes two artificial neural network (ANN) structures to enhance distance measurement accuracy using experimental data from a laser sensor. Both ANN structures are synthesized using the Multi-Layer Perceptron (MLP) model with different activation functions. The first structure utilizes the Rectified Linear Unit (ReLU), while the second employs the Hyperbolic Tangent function. Each ANN receives two inputs: the data has been gotten from laser sensor type VL53L0X and real measurements. The laser sensor is sensitive only to two colors (white and gray); as such, the gray and white obstacles are used. The ANNs are trained based on a dataset of 600 samples, with 80 percent for training, 20 percent for validation, and 20 percent for the testing. Measurement accuracy is evaluated with a Mean Squared Error (MSE) metric. The results based on the ReLU activation function showed higher accuracy than those with Tanh for the two colored objects. An ANN structure based on the ReLU function can be trained efficiently and accurately, making it suitable for robotic applications.

Zitieren

APA

Khalid, K. S., Khaleel, R. Z., Khaleel, H. Z., & Humaidi, A. J. (2026). Proposing Two Neural Networks Structures for Enhancing Laser Sensor Distance Accuracy. Tikrit Journal of Engineering Sciences, 33(2), 1-11. https://doi.org/10.25130/tjes.33.2.5

Vancouver

Khalid KS, Khaleel RZ, Khaleel HZ, Humaidi AJ. Proposing Two Neural Networks Structures for Enhancing Laser Sensor Distance Accuracy. Tikrit Journal of Engineering Sciences. 2026;33(2):1-11. doi:10.25130/tjes.33.2.5

BibTeX
@article{Khalid2026,
  author = {Khalid, Karam Samir and Khaleel, Rand Zuhair and Khaleel, Hind Zuhair and Humaidi, Amjad Jaleel},
  title = {Proposing Two Neural Networks Structures for Enhancing Laser Sensor Distance Accuracy},
  journal = {Tikrit Journal of Engineering Sciences},
  year = {2026},
  volume = {33},
  number = {2},
  pages = {1--11},
  doi = {10.25130/tjes.33.2.5}
}
Kostenlos starten →