info@ijrst.com

+919555269393

Track Article

Upload Article

IJRST
About
About IJRST Editorial Board Reviewers Advisory Board Call for Papers Subject Areas Indexing
For Author
Peer Review Guidelines Instruction for Authors Submission Process Manuscript Withdrawal Article Processing Charge Invitation for Reviewers Review Process Article Workflow
Policies
Open Access Policy Review Policy Publication Ethics Article Correction Policy Plagiarism Policy Conflict of Interest Policy
Archive
Past Issues Current Issue
Download
Paper Template Copyright Form Cover Page AIU Prospectus
Certificate FAQ Contact Upload Article Track Article

Details

Artificial Intelligence in Embedded Systems: A Review of Techniques, Applications, and Challenges

Sarmad Hamad Ibrahim Alfarag

Electrical Engineering Department, Wasit University, Republic of Iraq

8-24

Vol: 15, Issue: 3, 2025

Receiving Date: 2025-07-11 Acceptance Date:

2025-08-03

Publication Date:

2025-08-12

Download PDF

http://doi.org/10.37648/ijrst.v15i03.002

Abstract

The convergence of Artificial Intelligence (AI) and embedded systems has revolutionized modern electronic engineering, enabling intelligent functionalities in devices with constrained power, memory, and processing resources. This review explores the evolution, techniques, hardware platforms, and application domains of embedded AI, focusing on advancements such as TinyML, federated learning, and hybrid models. It further categorizes the landscape of AI-capable embedded hardware, including microcontrollers, SoCs, FPGAs, ASICs, and modular AI accelerators. Real-world deployments across agriculture, healthcare, automotive, IIoT, robotics, and smart cities are discussed, with emphasis on privacy-aware, real-time, and energy-efficient implementations. The paper also outlines critical challenges such as computational limits, latency, model updates, and security risks. Lastly, it highlights emerging trends including neuromorphic computing, self-learning models, cross-platform ML deployment, and hardware-algorithm co-design, offering a forward-looking perspective on the future of AI in embedded applications.

Keywords: Embedded artificial intelligence; Edge AI hardware; TinyML in embedded systems; AI-enabled microcontrollers; Low-power AI design; Real-time AI processing; Neuromorphic computing

References

  1. Abdi, H., et al. (2025). Emerging AI-enabled microcontrollers and neuromorphic chips. IEEE Access, 13, 99324– 99335.
  2. Abhay, A., & Sutar, S. (2025). AI in precision farming: An embedded approach. Agricultural AI Journal, 12(3), 88– 97.
  3. Abubakar, M., et al. (2025). IIoT: Embedded systems, TinyML, and federated learning. Journal of Computing and Biomedical Informatics.
  4. Ahmed, K., & Mehta, V. (2024). Self-adaptive embedded sensors for rotating machinery. Smart Manufacturing Letters, 9(3), 129–136.
  5. Al-Abdullah, S., & Park, M. (2024). Wearable AI systems for cardiovascular monitoring. Sensors and Actuators B: Chemical, 438, 135861. https://doi.org/10.1016/j.snb.2024.135861
  6. Aljundi, R., et al. (2021). Tiny continual learning for embedded AI. NeurIPS Workshop on Lifelong Learning
  7. Arduino AG. (2024). Portenta Vision Shield: Enabling AI at the edge. Arduino Developer Docs.
  8. Banbury, C., et al. (2021). Benchmarking TinyML systems: Challenges and direction. arXiv preprint. https://arxiv.org/abs/2102.05278
  9. Banbury, C., et al. (2023). TinyMLPerf: Benchmarking embedded ML systems. arXiv preprint. https://arxiv.org/abs/2303.10123
  10. Blazevic, R., Maaß, F. L., Kofler, C., & Veledar, O. (2025). RaspiCar: Ensuring safe real-time AI-based control for embedded autonomous systems. In Lecture Notes in Computer Science. Springer. https://doi.org/10.1007/978-3-032- 04291-0_10
  11. Chakranarayan, V., Shaker, R. J., Hussain, F., & Jaber, F. A. (2025). Safeguarding brand and platform credibility through AI-based multi-model fake profile detection. Future Internet, 17(9), 391. https://doi.org/10.3390/fi17090391
  12. Chen, R., & Tan, E. (2023). Driver state detection using TinyCNNs. Embedded AI Journal, 5(1), 17–29Chen, R., & Tan, E. (2023). Driver state detection using TinyCNNs. Embedded AI Journal, 5(1), 17–29Chen, R., & Tan, E. (2023). Driver state detection using TinyCNNs. Embedded AI Journal, 5(1), 17–29
  13. Chen, T., et al. (2024). Hybrid federated reinforcement learning in embedded robotics. Robotics and Autonomous Systems
  14. Chen, Y., et al. (2022). A survey on FPGA-based AI inference acceleration. ACM Computing Surveys, 55(1), 1– 35. https://doi.org/10.1145/3485126
  15. Davies, M., et al. (2021). Advancing neuromorphic computing with Intel Loihi. Nature Electronics, 4(5), 291– 302. https://doi.org/10.1038/s41928-021-00577-x
  16. Han, S., Mao, H., & Dally, W. J. (2016). Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding. International Conference on Learning Representations (ICLR).
  17. Han, S., et al. (2022). Model compression for edge AI: Techniques and tools. IEEE Transactions on Neural Networks and Learning Systems, 33(5), 1993–2005. https://doi.org/10.1109/TNNLS.2021.3106666
  18. Han, S., et al. (2023). Designing accelerators for TinyML: Challenges and directions. IEEE Design & Test, 40(2), 38– 49. https://doi.org/10.1109/MDAT.2022.3224853
  19. Hussain, M., et al. (2023). Real-time cardiac monitoring using low-power wearables. IEEE Sensors Journal, 23(4), 5056–5063. https://doi.org/10.1109/JSEN.2023.3236123
  20. Jeong, S., & Gao, X. (2023). Timing predictability in embedded RTOS for edge AI. *Real-Time Systems Journal, 59*(1), 33–50. https://doi.org/10.1007/s11241-022-09397-4
  21. Kairouz, P., et al. (2021). Advances and open problems in federated learning. Foundations and Trends® in Machine Learning, 14(1–2), 1–210. https://doi.org/10.1561/2200000083
  22. Kargar, A., et al. (2025). Concept drift mitigation on resource-constrained IoT devices via self-learning. IEEE Sensors Conference.
  23. Kargar, A., et al. (2025). Self-adaptive AI for concept drift in edge devices. IEEE Sensors Conference
  24. Kargar, A., Zorbas, D., & Gaffney, M. (2025). Self-learning IoT devices for edge AI. IEEE Sensors Conference.
  25. Kim, M., et al. (2024). Latency-aware AI in robotics. Robotics and Autonomous Systems, 155, 104189. https://doi.org/10.1016/j.robot.2022.104189
  26. Kneron Inc. (2023). KL520 Edge AI SoC for smart devices. Kneron Whitepaper
  27. Kumar, A., & Gupta, D. (2024). AI-powered edge systems: A new era in embedded intelligence. IEEE Internet of Things Magazine, 7(1), 22–30.
  28. Kumar, A., et al. (2025). Deploying TinyML models in smart farming devices. IEEE Access, 13, 88745– 88755. https://doi.org/10.1109/ACCESS.2025.3387612
  29. Kwon, H., et al. (2024). Memory-constrained deep learning models for IoT. ACM Journal on Emerging Technologies, 19(1), 21–32. https://doi.org/10.1145/3582567
  30. Lee, J., et al. (2023). Memory- and compute-aware AI for embedded devices. IEEE Embedded Systems Letters, 15(2), 89–96. https://doi.org/10.1109/LES.2022.3224532
  31. Lee, S. J., et al. (2025). Industrial AI systems for predictive maintenance. Journal of Industrial Informatics, 21(2), 220–235
  32. Li, H., et al. (2021). Efficient deep neural network deployment on Huawei's Kirin SoC. IEEE Embedded Systems Letters, 13(2), 45–49. https://doi.org/10.1109/LES.2020.3024567
  33. Li, Y., & Huang, J. (2025). AI accelerator design for energy-efficient edge devices. IEEE Access, 13, 80211–80225
  34. Lin, S. H., Jhuang, Y. C., Zha, C., et al. (2025). Spatiotemporal gesture recognition system based on landmarks on Jetson Nano. In Edge AI and ML Systems. Google Books.
  35. Lu, Q., et al. (2025). Flash-backed deep learning in embedded systems. IEEE Design & Test, 42(2), 45– 53. https://doi.org/10.1109/MDAT.2024.3387612
  36. Luo, C., Tang, H., & Li, S. (2025). Improved YOLOv11s for fault detection in embedded systems. IEEE Transactions on AI Systems.
  37. McMahan, H. B., et al. (2017). Communication-efficient learning of deep networks from decentralized data. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS).
  38. Mnih, V., et al. (2015). Human-level control through deep reinforcement learning. Nature, 518(7540), 529– 533. https://doi.org/10.1038/nature14236
  39. Montes-Rivera, M., Tavarez-Delgado, J. R., et al. (2025). Multi-objective assistant for control designing with overshoot suppression using genetic algorithms. ResearchGate. https://www.researchgate.net/publication/395022633
  40. Morales, E., et al. (2024). Ubiquitous embedded AI: The future of smart systems. Journal of Artificial Intelligence and Embedded Applications, 9(3), 141–155.
  41. Nardi, D., et al. (2023). Cross-platform deep learning with compiler-aware NAS. ACM Transactions on Embedded Computing Systems, 22(1), 11. https://doi.org/10.1145/3582567
  42. Papernot, N., & McDaniel, P. (2022). Security and trust in embedded AI systems. IEEE Security & Privacy, 20(6), 46–54. https://doi.org/10.1109/MSEC.2022.3188955
  43. Park, J., et al. (2022). Energy profiling of AI workloads on IoT devices. IEEE Transactions on Sustainable Computing, 7(3), 512–521. https://doi.org/10.1109/TSUSC.2022.3163226
  44. Patel, A., et al. (2023). Lightweight cryptography for embedded AI systems. Sensors and Secure Computing, 11(2), 77–91.
  45. Patel, D., et al. (2024). Federated machine learning for privacy-preserving IIoT. Journal of Edge Computing, 12(1), 45–59.
  46. Pazhani, A. A. J., & Vinodh, K. A. (2025). AI-based ULP microprocessors and microcontrollers. In Self-powered AIoT systems. Taylor & Francis. https://doi.org/10.1201/9781032684000-11
  47. Prajapati, D. (2024). Development of embedded AI applications & toolchain. Institute of Technology
  48. Rani, P., Sehrawat, H., Kaur, A., & Damaševičius, R. (2025). Emergent defenders: Generative AI's role in safeguarding IoT ecosystems. Cluster Computing. https://doi.org/10.1007/s10586-025-05126-1
  49. Raspberry Pi Foundation. (2023). Using the Neural Compute Stick with RPi4. Raspberry Pi Blog
  50. Razack, R. K., Poovadichalil, N. M., & Sadasivuni, K. K. (2025). Toward autonomous medicine: A review of biomedical energy harvesting and wearable sensing systems. Nano Energy, 108452. https://doi.org/10.1016/j.nanoen.2025.108452
  51. Roy, K., et al. (2022). Spiking neural networks for energy-efficient embedded inference. IEEE Transactions on Neural Networks and Learning Systems, 33(9), 7890–7902. https://doi.org/10.1109/TNNLS.2021.3085421
  52. Russo, G., et al. (2023). Edge AI for urban monitoring using low-cost devices. Smart Cities and Infrastructure, 10(4), 301–315.
  53. Sattler, F., Müller, K., et al. (2022). Clustered federated learning in non-IID settings. IEEE Transactions on Neural Networks and Learning Systems, 33(11), 6123–6135. https://doi.org/10.1109/TNNLS.2021.3073553
  54. Serrano-Gotarredona, T., et al. (2023). Event-based AI: From TrueNorth to BrainChip Akida. Frontiers in Neuroscience, 17, 1020183. https://doi.org/10.3389/fnins.2023.1020183
  55. Shokri, R., & Shmatikov, V. (2017). Privacy-preserving deep learning. Proceedings of the 24th ACM Conference on Computer and Communications Security (CCS).
  56. Singh, P., & Rani, R. (2023). Advancements in AI accelerators for edge devices. ACM Transactions on Embedded Computing Systems, 22(2), 15. https://doi.org/10.1145/3582567
  57. Sipola, T., Kokkonen, T., & Alatalo, J. (2022). Artificial intelligence in the IoT era: A review of edge AI hardware and software. IEEE Conference on Emerging Technologies. https://doi.org/10.1109/ET.2022.9770931
  58. Supriadi, C., Wahyudi, W., & Priyadi, A. (2025). Decentralized AI on the edge: Implementing federated learning for predictive maintenance in industrial IoT systems. Journal of Technology and IoT Management.
  59. Sze, V., et al. (2022). Hardware-algorithm co-design for efficient AI inference. Proceedings of the IEEE, 110(1), 68– 91. https://doi.org/10.1109/JPROC.2021.3122294
  60. Torres, J., et al. (2025). AI-enabled UAV systems for precision agriculture. IEEE Transactions on Robotics, 41(5), 1214–1228. https://doi.org/10.1109/TRO.2025.3387612
  61. Wang, J., & Kim, H. (2025). Cross-platform compilers for embedded AI: Challenges and tools. IEEE Embedded Systems Letters, 17(2), 88–96. https://doi.org/10.1109/LES.2024.3387612
  62. Wang, Y., & Zhao, L. (2024). Deep learning models for ADAS on Jetson Xavier. IEEE Transactions on Intelligent Vehicles, 9(1), 71–80. https://doi.org/10.1109/TIV.2023.3236123
  63. Wang, Y., et al. (2023). TinyRL: Resource-aware reinforcement learning for embedded devices. ACM Transactions on Embedded Computing Systems, 22(3), 1–25. https://doi.org/10.1145/3582567
  64. Warden, P., & Situnayake, D. (2019). TinyML: Machine learning with TensorFlow Lite on Arduino and ultra-lowpower microcontrollers. O'Reilly Media.
  65. Wu, J., et al. (2024). Real-time learning on tiny devices: A feasibility study. IEEE Internet of Things Journal, 11(1), 241–252. https://doi.org/10.1109/JIOT.2023.3236123
  66. Yin, Y., et al. (2024). Efficient quantized inference for low-power MCUs. ACM Transactions on Embedded Computing Systems, 23(4), 115. https://doi.org/10.1145/3582567
  67. Zhang, X., et al. (2024). TVM: An end-to-end optimizing compiler for deep learning. Communications of the ACM, 67(3), 78–87. https://doi.org/10.1145/3582567
  68. Zhang, Y., et al. (2022). Federated learning for on-device healthcare AI. ACM Transactions on Computing for Healthcare, 3(2), 1–21. https://doi.org/10.1145/3506718Zhang, Y., et al. (2022). Federated learning for on-device healthcare AI. ACM Transactions on Computing for Healthcare, 3(2), 1–21. https://doi.org/10.1145/3506718Zhang, Y., et al. (2022). Federated learning for on-device healthcare AI. ACM Transactions on Computing for Healthcare, 3(2), 1–21. https://doi.org/10.1145/3506718
  69. Zhang, Y., et al. (2024). Federated and continual learning in embedded AI. IEEE Transactions on Emerging Topics in Computing, 12(1), 19–34. https://doi.org/10.1109/TETC.2023.3236123
  70. Zhao, T., et al. (2024). Model updating in edge AI. IEEE Internet of Things Journal, 11(3), 2125– 2137. https://doi.org/10.1109/JIOT.2023.3236123
Back

Disclaimer: Indexing of published papers is subject to the evaluation and acceptance criteria of the respective indexing agencies. While we strive to maintain high academic and editorial standards, International Journal of Research in Science and Technology does not guarantee the indexing of any published paper. Acceptance and inclusion in indexing databases are determined by the quality, originality, and relevance of the paper, and are at the sole discretion of the indexing bodies.

We are one of the best in the field of watches and we take care of the needs of our customers and produce replica watches of very good quality as per their demands.

Our Head Office

Aryavart International University
Mission Tilla, Dharmanagar, North Tripura, India

+91-9555269393

info@ijrst.com

Quick Links

Plagiarism Policy Current Issue Certificate Reviewers FAQs Our Journals

Indexing

© IJRST, All Right Reserved.

Designed By Krrypto