Raspberry Pi 5 and Its Applications in ML Models

Abstract

Raspberry Pi 5 is the newest development board in the Raspberry Pi series. This paper describes the architecture, features, interfacing capabilities, performances and its appli- cations in the machine learning models. The paper describes the hardware advantages necessary for efficently training and running ML models. This document also discusses various possibilities of ML models that can be trained in these boards. The paper concludes with the extensive possible applications of Raspberry Pi 5 board in IoT and general electronics. This advancements may provide a platform for the innovations of many newer generation ideas and projects.

Introduction

Technological advancements have significantly impacted human life, and the Raspberry Pi series is a key innovation enabling real-world applications of computing and machine learning (ML). The Raspberry Pi 5, launched in 2023, is a compact mini-computer (85 × 49 mm) running the Debian-based “Bookworm” OS. It features a 64-bit quad-core ARM Cortex-A76 processor (BCM2712), VideoCore VII GPU, dual 4Kp60 HDMI outputs, LPDDR4X memory (2–16 GB), dual-band Wi-Fi, Bluetooth 5.0/BLE, USB 3.0/2.0 ports, Gigabit Ethernet with PoE+, PCIe 2.0 interface, microSD support, MIPI camera/display transceivers, and a real-time clock (RTC). Power is supplied via USB-C (5V, 5A), supporting demanding workloads, peripherals, and ML applications. A standard 40-pin GPIO header enables extensive interfacing for sensors, motor control, audio, and communications.

The Raspberry Pi 5’s hardware provides a robust platform for IoT and edge ML applications, including computer vision, deep learning, and AI. Literature reviews highlight successful implementations of computer vision algorithms on Raspberry Pi devices, such as object counting and automated depalletizing using SIFT, ORB, Haar cascades, and pattern-matching methods. The board’s enhanced processing power, memory bandwidth, and peripheral support make it suitable for training, testing, and deploying ML models in real-world scenarios.

Conclusion

The study shows a large scope for the use and development of Raspberry Pi 5 for better utilization of ML models. The Raspberry Pi 5 is a small package with a lot of functionalities compacted into a module that is cost-effective and very useful especially when used for prototype development. The ML modeling capabilities can be extended for the integration of AI into the new upcoming technology. The balance of advan- tages and disadvantages for the development is crucial when considering for producing the prototype. The ML development is a being of AI integration and Raspberry Pi 5 is a important part for its integration into IoT and networking systems.

References

[1] “Processors - Raspberry Pi documentation.” https://www.raspberrypi.com/documentation/computers/processors.html [2] S. Imtiaz, “The Raspberry Pi 5: a huge improvement from its prede- cessor,” Nov. 2023. [Online]. Available: https://elektormagazine.s3.eu- central-1.amazonaws.com/private/ep/ideas/230635/articles/230635-01-1.pdf?X-Amz-Content-Sha256=UNSIGNED-PAYLOAD&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz- Credential=AKIATBZISSD62LEER6BQ%2F20251101%2Feu-central-1%2Fs3%2Faws4request&X-Amz-Date=20251101T172541Z&X- Amz-SignedHeaders=host&X-Amz-Expires=600&X-Amz- Signature=dce1fbe00d2a31412fc83cb4922fc9b1df6123b5a06a885ec7- eea1c180833d17 [3] “Fundamentals of Wi-Fi 6: Capacity is the new metric,” Cisco. Jul. 10, 2025. [Online]. Available: https://www.cisco.com/site/us/en/learn/topics/networking/what-is-802-11ac.html [4] Cypress Semiconductor Corporation, “CYW43455Single- Chip5G WiFi IEEE 802.11n/ac MAC/Baseband/Radio with Integrated Bluetooth 5.0,” Cypress Semiconductor Corporation, report, Mar. 2019. [Online]. Available: https://www.infineon.com/assets/row/public/documents/30/49/infineon- cyw43455-datasheet-en.pdf?fileId=8ac78c8c7d0d8da4017d0ee226686889 [5] E. Notes, “Gigabit Ethernet & 1000Base T Electronics Notes.” https://www.electronics-notes.com/articles/connectivity/ethernet-ieee- 802-3/gigabit-ethernet-1ge-1000-base-t.php [6] Texas Instruments, KeyStone Architecture Peripheral Component In- terconnect Express (PCIE) User Guide. 2013. [Online]. Available: https://www.ti.com/lit/ug/sprugs6d/sprugs6d.pdf?ts=1761989908564 [7] Abdulhamid, Mohanad & Al-Rawi, Muhanned. (2020). “COMPUTER VISION BASED ON RASPBERRY PI SYSTEM”. 10.23743/acs-2020- 31. [8] Greco, Danilo, Majid Fasihiany, Ali Varasteh Ranjbar, Francesco Ma- sulli, Stefano Rovetta, and Alberto Cabri. 2024.“Computer Vision Al- gorithms on a Raspberry Pi 4 for Automated Depalletizing” Algorithms 17, no. 8: 363. https://doi.org/10.3390/a17080363 [9] Hobbs, Thomas & Ali, Anwar. (2025). Smart Home Control Using Real- Time Hand Gesture Recognition and Artificial Intelligence on Raspberry Pi 5. Electronics. 14. 3976. 10.3390/electronics14203976. [10] James, Nicholas, Lee-Yeng Ong, and Meng-Chew Leow. 2022.“Explor- ing Distributed Deep Learning Inference Using Raspberry Pi Spark Clus- ter” Future Internet 14, no. 8: 220. https://doi.org/10.3390/fi14080220 [11] Sanjay S K, Supriya G K, Sachidanand M, Harshitha G, Prof. Akshatha B G, “Deep Learning Approach to Detect Fake Video on Raspberry PI”, ISSN: 2321-9653, Volume.12, Issue VII, https://doi.org/10.22214/ijraset.2024.60781 [12] Shafia Firdose,Shiva Kumar B K,Dinesh G, ”RASPBERRY PI-BASED DEEP LEARNING FACE RECOGNITION SECURITY SYSTEM”, International Journal of Creative Research Thoughts (IJCRT), ISSN:2320- 2882, Volume.9, Issue 2, pp.5498-5507, February 2021, Available at: http://www.ijcrt.org/papers/IJCRT2102668.pdf [13] Deniz, Mehmet & Bogrekci, Ismail & Demircioglu, Pinar. (2025). Real-Time Detection of Hole-Type Defects on Industrial Compo- nents Using Raspberry Pi 5. Applied System Innovation. 8. 1-26. 10.3390/asi8040089. [14] Tigadi, Arun & rodrigues, kevin. (2023). “Study on Building a Raspberry Pi AI system: Tools and Techniques”. 10. 476-482. [15] Kanade, Prakash & Ghatwal, Shama & Kanade, Sunay & Ku- mar, Divya. (2025). “AI-Powered Robotic Lawn Mower with Au- tonomous Navigation and Control Using Raspberry Pi”. 2395-0056. https://doi.org/10.5281/zenodo.17170792 [16] Steadman, P., Jenkins, P., Rathore, R. S., & Hewage, C. (2024). Challenges in Implementing Artificial Intelligence on the Raspberry Pi 4, 5 and 5 with AI HAT. In N. Naik, P. Grace, P. Jenkins, & S. Prajapat (Eds.), Contributions Presented at The International Conference on Computing, Communication, Cybersecurity and AI - The C3AI 2024 (pp. 147-157). (Lecture Notes in Networks and Systems; Vol. 884 LNNS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-74443-3 8 [17] Sudha Ellison Mathe, Hari Kishan Kondaveeti, Suseela Vappangi, Sunny Dayal Vanambathina, Nandeesh Kumar Kumaravelu, “A comprehensive review on applications of Raspberry Pi”, Com- puter Science Review, Volume 52, 2024, 100636, ISSN 1574-0137, https://doi.org/10.1016/j.cosrev.2024.100636.

Copyright

Copyright © 2025 Chetan G Kotangale, Prof. Vishal Ambhore, Dr. Pravin G Gawande. 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.