A Hardware/Software System for Obstacle Detection in Urban Roads

• Allicia R. dos Santos IFES
• Julia S. Barreiros IFES
• João V. S. da Silva IFES
• Amanda B. Ferreira IFES
• Richard J. M. G. Tello IFES

DOI: https://doi.org/10.5753/eries.2025.16039

Abstract

This paper presents the development of an embedded hardware and software system for obstacle detection on roads. The system consists of a 4WD mobile robot equipped with an ESP32 board and an MPU6050 inertial sensor. Scaled obstacles, produced by 3D printing, were tested in linear and rectangular scenarios. The data collected by the inertial sensor were transmitted via Wi-Fi for external processing. After preprocessing and segmentation of the temporal signals, recurrent neural networks (LSTM, Bi-LSTM, and GRU) were employed for analysis. The results indicated that the GRU network achieved the best performance, reaching an average accuracy above 98% in both evaluated scenarios. These findings demonstrate the feasibility of the system for identifying irregularities in controlled environments.

References

Cho, K., Gulcehre, C., Bougares, F., Schwenk, H., and Bengio, Y. (2014). Learning phrase representations using rnn encoder-decoder for statistical machine translation. arXiv preprint arXiv:1406.1078.

Confederação Nacional do Transporte (2024). Pesquisa CNT de Rodovias 2024. CNT; SEST SENAT; ITL, Brasília. ISBN 978-85-68865-23-1.

De Zoysa, K., Keppitiyagama, C., and Weerathunga, S. (2007). A public transport system based sensor network for road surface condition monitoring. page 9.

Eriksson, J., Girod, L., Hull, B., Newton, R., Madden, S., and Balakrishnan, H. (2008). The pothole patrol: Using a mobile sensor network for road surface monitoring. In Proceedings of the 6th International Conference on Mobile Systems, Applications, and Services (MobiSys ’08).

Hochreiter, S. and Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8):1735–1780.

Mednis, A., Strazdins, G., Zviedris, R., Kanonirs, G., and Selavo, L. (2011). Real time pothole detection using android smartphones with accelerometers. pages 1 – 6.

Rodas, Q. (2018). Novas tecnologias: carros atuais têm até 100 sensores a bordo. Revista Quatro Rodas. Acesso em: 8 ago. 2025.

Rosca, C., Stancu, A., and Gortoescu, I.-A. (2025). Advanced sensor integration and ai architectures for next-generation traffic navigation. Applied Sciences, 15:4301.

Schuster, M. and Paliwal, K. K. (1997). Bidirectional recurrent neural networks. IEEE Transactions on Signal Processing, 45(11):2673–2681.

Ye, Z., Wei, Y., Yang, S., Li, P., Yang, F., Yang, B., and Wang, L. (2024). Iot-enhanced smart road infrastructure systems for comprehensive real-time monitoring. Internet of Things and Cyber-Physical Systems, 4:235–249.

Zareei, M., Castañeda, C. A. L., Alanazi, F., Granda, F., and Díaz, J. A. P. (2025). Machine learning model for road anomaly detection using smartphone accelerometer data. IEEE Access, pages 1–1.