AI-Enabled Sustainable Precision Farming Using Raspberry Pi and Coral Edge TPU with Cloud-Based Model Training

Abstract

Precision Agriculture requires intelligent, adaptable and efficient technology that is capable of making decisions in agriculture at the right time, based on the condition of the plants and the environment. This paper proposes the development of an AI-based sustainable precision farming system based on computer vision, cloud based model training, Raspberry Pi, Coral Edge TPU and the Internet of Things (IoT) sensing. The proposed architecture\'s sensing layer, edge processing layer, cloud intelligence layer, and application layer enables a complete data processing pipeline from the field to real time decision support. Crop photos are collected for AI condition classification and agricultural conditions such as soil moisture, temperature, humidity and light intensity are collected by built-in sensors. After the above preprocessing, normalisation, image resizing and augmentation, the dataset is divided into training, validation and test sets. The trained deep learning models in a cloud environment are then quantized and converted to be optimized for the edge, following this step. Following this step, the trained deep learning models are quantized and converted for optimization on edge. The optimized model achieves an exemplary accuracy of 94.82%, precision of 94.47%, recall of 94.05% and F1-score of 94.26%. Deploying with the Coral Edge TPU reduces inference latency to approximately 38ms compared to Raspberry Pi-only processing, increases processing throughput to 26.3 pictures per second and reduces CPU usage. The edge-assisted architecture reduces the need for constant connection to the cloud, giving local decision making a leg up. Illustrative sustainability study estimates that condition based monitoring can save 22% of the irrigation water used every day. The findings indicated that the introduction of the edge AI and cloud intelligence in precision agricultural systems that are responsive, scalable, and resource-efficient is technically possible.

Introduction

The text presents a research study on developing a sustainable AI-based precision farming system that combines IoT sensors, Raspberry Pi, Coral Edge TPU, cloud computing, machine learning, and deep learning to improve agricultural monitoring and decision-making.

Traditional farming practices often apply water, fertilizers, pesticides, and other resources uniformly, even though crop and soil conditions can vary significantly. This can lead to resource wastage, higher costs, soil degradation, excessive chemical use, and inefficient water consumption. Precision agriculture addresses these problems by using sensors, automation, data analysis, and AI to provide resources according to actual crop and environmental needs.

Key Technologies

The proposed system combines several technologies:

• IoT sensors: Continuously monitor parameters such as soil moisture, temperature, humidity, and light intensity.

• Raspberry Pi: Acts as a low-cost local gateway and processing unit that collects sensor and camera data.

• Coral Edge TPU: Accelerates AI inference directly on the farm, allowing rapid analysis without depending completely on the cloud.

• Cloud computing: Provides the computational resources needed for dataset management, model training, validation, optimization, and model updates.

• AI/ML/DL: Analyze sensor and image data to identify crop health problems, diseases, pests, environmental conditions, and other agricultural issues.

• Computer vision: Uses crop and leaf images to detect diseases, stress, and other visible problems.

Why Edge Computing Is Important

A major problem with fully cloud-based agricultural systems is that continuously sending sensor and image data to the this problem by performing initial analysis locally. The Raspberry Pi and Coral Edge TPU process data at the farm, while the cloud performs computationally intensive model training. Only cloud requires high bandwidth, reliable Internet connectivity, and additional processing time. This can be particularly difficult in rural and remote agricultural areas.

The proposed edge-cloud architecture addresses this problem by performing initial analysis locally. The Raspberry Pi and Coral Edge TPU process data at the farm, while the cloud performs computationally intensive model training. Only important results or selected data need to be transferred to the cloud. This approach can provide:

• Faster real-time decisions

• Reduced bandwidth and communication requirements

• Lower dependence on Internet connectivity

• More efficient use of computing resources

• Energy-efficient local AI processing

• Easier model updates through cloud-based retraining

Research Objectives

The research aims to:

1. Develop an AI-enabled precision farming system using cloud-based model training, IoT sensors, Raspberry Pi, and Coral Edge TPU.

2. Evaluate the system based on AI prediction accuracy, edge inference performance, resource consumption, processing efficiency, and agricultural sustainability.

Scope of the Research

The system focuses on real-time monitoring of:

• Crop health

• Temperature

• Humidity

• Soil moisture

• Other environmental and agricultural conditions

Machine learning and deep learning are used for crop health assessment, anomaly detection, prediction, and agricultural decision support. The cloud is responsible for training and optimizing models, while the Coral Edge TPU performs efficient local inference.

Related Work

Previous studies demonstrate that AI, IoT, machine learning, deep learning, and edge computing can support many agricultural applications, including:

• Crop yield prediction

• Plant disease detection

• Pest detection

• Irrigation management

• Weed detection

• Plant phenotyping

• Environmental monitoring

• Soil and water management

• Automated agricultural decision support

Research also shows that combining sensor data with machine-learning models can improve irrigation and resource management. However, existing systems face challenges such as connectivity, scalability, data management, computational limitations, and security.

Research Gap

The main research gap identified is that many existing precision-farming systems rely heavily on cloud computing. This can cause:

• Communication delays

• High bandwidth consumption

• Dependence on stable Internet connections

• Reduced performance in remote areas

Although Raspberry Pi is affordable and suitable for agricultural IoT applications, it has limited computational power for complex deep-learning models. Few studies have specifically investigated the combination of Raspberry Pi + Coral Edge TPU + cloud-based model training for sustainable precision agriculture.

Therefore, the proposed research aims to develop an integrated edge-cloud architecture that provides real-time AI inference while reducing communication requirements and improving resource efficiency.

Proposed System Architecture

The proposed architecture consists of four major layers:

1. Sensing layer: IoT sensors and cameras collect soil, environmental, and crop information.

2. Edge-processing layer: Raspberry Pi collects and processes the data, while the Coral Edge TPU accelerates AI inference.

3. Cloud-intelligence layer: The cloud stores data and performs model training, validation, optimization, and updates.

4. Application/decision-support layer: The analyzed information is converted into useful agricultural recommendations and decisions.

Conclusion

In this paper, an AI powered sustainable precision farming solution involving cloud based model training, edge computing with a Raspberry Pi, IoT sensing, computer vision, and Coral Edge TPU acceleration is proposed. The architecture forms an integrated path for data collection, preprocessing, smart analysis and real time decision making in agriculture. According to the experimental investigation, the optimized edge model may significantly increase computing efficiency while maintaining good predictive performance, attaining 94.82% accuracy and a 94.26% F1-score. The results of the Coral Edge TPU acceleration reveal that dedicated edge hardware is a suitable solution for real-time applications in agriculture, as it reduces inference delay and CPU usage. The edge cloud architecture also eliminates the need for constant connectivity between the edge and the cloud because it allows for the selection of processed information transmitted to the infrastructure in the cloud. This allows for local inference and selective transmission of processed data to cloud infrastructure, further decreasing dependency on constant connectivity between the edge and the cloud. Moreover, condition based monitoring can be a component of precision agriculture\'s sustainability objectives and provide a technology foundation for more efficient irrigation and resource utilization. Therefore, the proposed solution addresses the integration of AI, IoT, and edge-cloud computing to boost the response, scalability, and operational efficiency of smart agriculture environments.

References

[1] Ayaz, M., Ammad-Uddin, M., Sharif, Z., Mansour, A., & Aggoune, E.-H. M. (2019). Internet-of-Things (IoT)-based smart agriculture: Toward making the fields talk. IEEE Access, 7, 129551–129583. https://doi.org/10.1109/ACCESS.2019.2932609 [2] Cambra Baseca, C., Sendra, S., Lloret, J., & Tomas, J. (2019). A smart decision system for digital farming. Agronomy, 9(5), 216. https://doi.org/10.3390/agronomy9050216 [3] Chung, H., Kim, J.-H., Ahn, J., Chung, Y., Kim, E., & Heo, W. (2026). Resource-constrained edge AI solution for real-time pest and disease detection in chili pepper fields. Agriculture, 16(2), 223. https://doi.org/10.3390/agriculture16020223 [4] Crane-Droesch, A. (2018). Machine learning methods for crop yield prediction and climate change impact assessment in agriculture. Environmental Research Letters, 13(11), 114003. https://doi.org/10.1088/1748-9326/aae159 [5] Devarajan, D., Allafi, R., Obayya, M., & Nemri, N. (2026). AI based real time disease diagnosis in plants using deep learning driven CNNs. Scientific Reports, 16(1), 4587. https://doi.org/10.1038/s41598-025-34681-1 [6] Elijah, O., Rahman, T. A., Orikumhi, I., Leow, C. Y., & Hindia, M. N. (2018). An overview of Internet of Things (IoT) and data analytics in agriculture: Benefits and challenges. IEEE Internet of Things Journal, 5(5), 3758–3773. https://doi.org/10.1109/JIOT.2018.2844296 [7] Goap, A., Sharma, D., Shukla, A. K., & Rama Krishna, C. (2018). An IoT based smart irrigation management system using machine learning and open source technologies. Computers and Electronics in Agriculture, 155, 41–49. https://doi.org/10.1016/j.compag.2018.09.040 [8] Jabed, M. A., & Murad, M. A. A. (2024). Crop yield prediction in agriculture: A comprehensive review of machine learning and deep learning approaches, with insights for future research and sustainability. Heliyon, 10(24), e40836. https://doi.org/10.1016/j.heliyon.2024.e40836 [9] Jeong, J. H., Resop, J. P., Mueller, N. D., Fleisher, D. H., Yun, K., Butler, E. E., Timlin, D. J., Shim, K.-M., Gerber, J. S., Reddy, V. R., & Kim, S.-H. (2016). Random forests for global and regional crop yield predictions. PLoS ONE, 11(6), e0156571. https://doi.org/10.1371/journal.pone.0156571 [10] Joice, A., Tufaique, T., Tazeen, H., Igathinathane, C., Zhang, Z., Whippo, C., Hendrickson, J. R., & Archer, D. (2025). Applications of Raspberry Pi for precision agriculture—A systematic review. Agriculture, 15(3), 227. https://doi.org/10.3390/agriculture15030227 [11] Kamilaris, A., & Prenafeta-Boldú, F. X. (2018). Deep learning in agriculture: A survey. Computers and Electronics in Agriculture, 147, 70–90. https://doi.org/10.1016/j.compag.2018.02.016 [12] Kamienski, C., Soininen, J.-P., Taumberger, M., Dantas, R., Toscano, A., Cinotti, T. S., Maia, R. F., & Torre Neto, A. (2019). Smart water management platform: IoT-based precision irrigation for agriculture. Sensors, 19(2), 276. https://doi.org/10.3390/s19020276 [13] Kanade, P. (2025). Raspberry Pi based automated garden sprinkler system with integrated weather, humidity, and soil moisture monitoring. International Research Journal of Engineering and Technology, 12(12), 728–736. https://doi.org/10.5281/ZENODO.21940827 [14] Kanade, P., Akhtar, M., David, F., & Kanade, S. (2021). Agricultural mobile robots in weed management and control. International Journal of Advanced Networking and Applications, 13(3), 5001–5006. [15] Kanade, P., & Ashwini, A. P. (2021). Smart agriculture robot for sowing seed. International Journal of Engineering Science and Computing, 11(1), 27563–27565. [16] Kanade, P., & Prasad, J. P. (2021a). Arduino based machine learning and IoT smart irrigation system. International Journal of Soft Computing and Engineering, 10(4), 1–5. [17] Kanade, P., & Prasad, J. P. (2021b). Machine learning techniques in plant conditions classification and observation. In Proceedings of the 2021 5th International Conference on Computing Methodologies and Communication (ICCMC) (pp. 729–734). IEEE. https://doi.org/10.1109/ICCMC51019.2021.9418386 [18] Khanna, A., & Kaur, S. (2019). Evolution of Internet of Things (IoT) and its significant impact in the field of precision agriculture. Computers and Electronics in Agriculture, 157, 218–231. https://doi.org/10.1016/j.compag.2018.12.039 [19] Kim, Y., Evans, R. G., & Iversen, W. M. (2008). Remote sensing and control of an irrigation system using a distributed wireless sensor network. IEEE Transactions on Instrumentation and Measurement, 57(7), 1379–1387. https://doi.org/10.1109/TIM.2008.917198 [20] Liakos, K. G., Busato, P., Moshou, D., Pearson, S., & Bochtis, D. (2018). Machine learning in agriculture: A review. Sensors, 18(8), 2674. https://doi.org/10.3390/s18082674 [21] Ray, P. P. (2017). Internet of Things for smart agriculture: Technologies, practices and future direction. Journal of Ambient Intelligence and Smart Environments, 9(4), 395–420. https://doi.org/10.3233/AIS-170440 [22] Saiz-Rubio, V., & Rovira-Más, F. (2020). From smart farming towards Agriculture 5.0: A review on crop data management. Agronomy, 10(2), 207. https://doi.org/10.3390/agronomy10020207 [23] Tzounis, A., Katsoulas, N., Bartzanas, T., & Kittas, C. (2017). Internet of Things in agriculture, recent advances and future challenges. Biosystems Engineering, 164, 31–48. https://doi.org/10.1016/j.biosystemseng.2017.09.007 [24] Van Klompenburg, T., Kassahun, A., & Catal, C. (2020). Crop yield prediction using machine learning: A systematic literature review. Computers and Electronics in Agriculture, 177, 105709. https://doi.org/10.1016/j.compag.2020.105709 [25] Veenadhari, S., Misra, B., & Singh, C. D. (2014). Machine learning approach for forecasting crop yield based on climatic parameters. In Proceedings of the 2014 International Conference on Computer Communication and Informatics (ICCCI) (pp. 1–5). IEEE. https://doi.org/10.1109/ICCCI.2014.6921718 [26] Wang, X., Huang, J., Feng, Q., & Yin, D. (2020). Winter wheat yield prediction at county level and uncertainty analysis in main wheat-producing regions of China with deep learning approaches. Remote Sensing, 12(11), 1744. https://doi.org/10.3390/rs12111744 [27] Wolfert, S., Ge, L., Verdouw, C., & Bogaardt, M.-J. (2017). Big data in smart farming: A review. Agricultural Systems, 153, 69–80. https://doi.org/10.1016/j.agsy.2017.01.023 [28] Xu, J., Gu, B., & Tian, G. (2022). Review of agricultural IoT technology. Artificial Intelligence in Agriculture, 6, 10–22. https://doi.org/10.1016/j.aiia.2022.01.001

Copyright

Copyright © 2026 Prakash Kanade, Shetty Chirag, Sunay Kanade. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.