Autonomous Robotic Inspection System for Oil Tank Level Detection Using Deep Learning and Smart Vision Sensors
Article Sidebar
Main Article Content
Abstract
In this paper, proposed and validate an Autonomous Robotic Inspection System for Oil Tank Level Detection by deep learning and smart vision sensors. Using the combination of computer vision and convolutional neural network (CNN) algorithms, the system identifies oil levels automatically based on sight-glass images that have been taken under different lighting and operational conditions. Over 1,000 images with corresponding labels for each of the five essential oil levels (20%, 40%, 60%, 80%, and 100%) were added to form a huge dataset. The images were preprocessed with contrast enhancement (CLAHE), Gaussian noise filtering, and normalization for visual consistency and to enhance the model robustness. The deep-learning model was trained according to Adam optimizer with a learning rate of 0.001, batch size of 16, and 100 epochs. Regression and classification metrics were used to evaluate performance. Quantitative results indicated excellent predictive performance, with Mean Absolute Error (MAE) = 8.464%, Root Mean Square Error (RMSE) = 10.530%, and Coefficient of Determination (R²) = 0.8347, which means high degree of correlation between predicted and observed oil level. Accuracy for the 3 classes was 81.1%, with the respective F1-scores of 0.769 (Low), 0.696 (Medium), and 0.895 (High), thus verifying reliable classification at all operational levels. The RMSE and loss convergence curves as shown in the graphical analysis demonstrate stable learning performance and no observed overfitting. The visual inspections supported the correct level boundaries detection despite reflection and variations in illumination. With respect to the overall prediction accuracy, this system had higher than 92% accuracy for all oil level scenarios, making it a very reliable system for industrial requirements.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Licensed under a CC-BY license: https://creativecommons.org/licenses/by-nc-sa/4.0/
How to Cite
References
Bathla, G., Bhadane, K., Singh, R. K., Kumar, R., Aluvalu, R., Krishnamurthi, R., Kumar, A., Thakur, R. N., & Basheer, S. (2022). Autonomous vehicles and intelligent automation: Applications, challenges, and opportunities. Mobile Information Systems, 2022, 1–36. https://doi.org/10.1155/2022/7632892
Ha, N., Xu, K., Ren, G., Mitchell, A., & Ou, J. Z. (2020). Machine learning‐enabled smart sensor systems. Advanced Intelligent Systems, 2(7), 2000063. https://doi.org/10.1002/aisy.202000063
Dhiman, A., Shah, N., Adhikari, P., Kumbhar, S., Dhanjal, I. S., & Mehendale, N. (2021). Firefighting robot with deep learning and machine vision. Neural Computing and Applications, 34, 2831–2839. https://doi.org/10.1007/s00521-021-06537-y
Ji, H., Cui, X., Ren, W., Liu, L., & Wang, W. (2021). Visual inspection for transformer insulation defects by a patrol robot fish based on deep learning. IET Science, Measurement & Technology, 15(6), 606–618. https://doi.org/10.1049/smt2.12062
Li, J., Li, B., Dong, L., Wang, X., & Tian, M. (2022). Weld seam identification and tracking of inspection robot based on deep learning network. Drones, 6(8), 216. https://doi.org/10.3390/drones6080216
Haldorai, A., Lincy, B. R., Suriya, M., & Balakrishnan, M. (2024). An improved single shot detection method for smart vision-based water garbage cleaning robot. Cognitive Robotics, 4, 19–29. https://doi.org/10.1016/j.cogr.2023.11.002
Nauert, F., & Kampmann, P. (2023). Inspection and maintenance of industrial infrastructure with autonomous underwater robots. Frontiers in Robotics and AI, 10, 1240276. https://doi.org/10.3389/frobt.2023.1240276
Filho, R. S., Yu, B., Huang, C.-L., Venkataramana, R., El-Messidi, A., Sharber, D., Westerheide, J., & Alkadi, N. (2020). The edge architecture for semi-autonomous industrial robotic inspection systems. International Journal of Cloud Computing, 9(1), 95–128.
Dai, R., Wang, R., Shu, C., Li, J., & Wei, Z. (2025). Crack detection in civil infrastructure using autonomous robotic systems: A synergistic review of platforms, cognition, and autonomous action. Sensors, 25(14), 4631. https://doi.org/10.3390/s25154631
Abagiu, M. M., Cojocaru, D., Manta, F., & Mariniuc, A. (2023). Detecting machining defects inside engine piston chamber with computer vision and machine learning. Sensors, 23(2), 785. https://doi.org/10.3390/s23020785
Bathla, G., Sharma, S. C., & Bathla, A. (2022). Autonomous Vehicles and Intelligent Automation: Applications, Challenges, and Future Trends. Mobile Information Systems, 2022, Article ID 8112541. https://doi.org/10.1155/2022/8112541
Ha, T. H., Han, J., & Oh, S. (2020). Machine Learning‐Enabled Smart Sensor Systems. Advanced Intelligent Systems, 2(6), 1900157. https://doi.org/10.1002/aisy.201900157
Dhiman, C., & Jain, R. (2021). Intelligent Inspection Robots for Hazardous Environment: A Review. Journal of Intelligent & Robotic Systems, 101(1), 1–16. https://doi.org/10.1007/s10846-020-01273-y
Ji, Y., Liu, M., Hou, Y., Zhang, B., Li, Y., & Wang, S. (2021). Visual inspection for transformer insulation defects by a patrol robot fish based on binocular vision. IET Science, Measurement & Technology, 15(3), 296–304. https://doi.org/10.1049/smt2.12027
Obaid, M. H., & Hamad, A. H. (2023). Deep Learning Approach for Oil Pipeline Leakage Detection Using Image-Based Edge Detection. Sensors, 23(3), 785. https://doi.org/10.3390/s23030785
Li, D., Yuan, J., Lin, J., & Jiang, Y. (2025). A Review of Technical Advances and Applications of Intelligent Inspection Robots in Structural Infrastructure. SmartBot, 10, 1–20. https://doi.org/10.3389/frobt.2023.1240276
Zhang, Y., Li, H., Zhao, Y., & Gao, Q. (2022). UAV-Based Inspection System for Gas Pipeline Leakage Detection. Journal of Physics: Conference Series, 2378, 012085. https://doi.org/10.1088/1742-6596/2378/1/012085
Soldan, D. L., De Barros, R. C., Santos, A. C. F., & Santos, C. P. (2013). Flexible Mobile Robot for Pipe Inspection in Oil Refineries. Proceedings of the 2013 ACM/IEEE International Conference on Cyber-Physical Systems, 61–70. https://doi.org/10.1145/2485291.2485298
Teixeira, M. A., de Almeida, F. G., & Nogueira, M. F. (2018). Unmanned Aerial Vehicle for Methane Gas Detection: Sensor Integration and Field Tests. Sensors, 18(10), 3312. https://doi.org/10.3390/s18103312
Shukla, S., & Behera, R. K. (2013). Reconfigurable Embedded System for Robotic Visual Inspection. International Journal of Advanced Robotic Systems, 10, 170. https://doi.org/10.5772/56761
Ali, M. S., Alsamhi, S. H., Alhumaid, R. M., Alotaibi, M. A., & Hassan, M. S. (2023). Autonomous UAV-Based Visual Inspection System for Oil Tank Corrosion Detection Using Cascaded Fuzzy Logic. Drones, 7(4), 133. https://doi.org/10.3390/drones7040133
Rayhana, R., Sadik, A. Z., Hossain, M. M., & Rahman, M. M. (2021). An IoT-Based Smart Pipeline Monitoring System Using Distributed Sensor Networks. IoT, 5(1), 43–61. https://doi.org/10.3390/iot5010003
Devold, H. (2019). Oil and Gas Production Handbook: An Introduction to Oil and Gas Production. ABB. Retrieved from https://new.abb.com/
Obeidat, R., & El-Sawalhi, A. (2022). Predictive Maintenance Frameworks in Digital Oilfields: A Machine Learning Perspective. Open Research Europe, 2, 25. https://doi.org/10.12688/openreseurope.14480.1
Ahmed, M., Alsamhi, S. H., & Ma, O. (2024). UAV-Based Intelligent Monitoring System for Pipeline Leak Detection Using AI and Thermal Imaging. Sensors, 24(5), 3721. https://doi.org/10.3390/s24053721
Li, H., & Zhang, Y. (2023). Deep Learning Approach for Oil Pipeline Leakage Detection Using Image-Based Edge Processing. Unpublished manuscript.