Low-Power AI Architectures for Energy-Efficient Industrial Automation

  • Authors

    • Dr. Harish Patel Professor, Gujarat University, India. Author
    • Dr. Geetha Ramasamy Associate Professor, Madurai Kamaraj University, India. Author

    DOI:

    https://doi.org/10.67228/3142788X/IJMLPA-2020PII9C8P

    Published 07-02-2020

  • Low-Power AI, Energy-Efficient Computing, Industrial Automation, Edge AI, TinyML, Neuromorphic Computing, AI Accelerators, Embedded Intelligence, Green AI, Predictive Maintenance

    Issue

    Vol. 3 No. 2 (2020)

    Section

    Articles

    How to Cite

    [1]
    H. Patel and G. Ramasamy, “Low-Power AI Architectures for Energy-Efficient Industrial Automation”, IJMLPA, vol. 3, no. 2, pp. 01–10, Jul. 2020, doi: 10.67228/3142788X/IJMLPA-2020PII9C8P.

  • Abstract

    Industrial automation is undergoing a significant transformation with the integration of Artificial Intelligence (AI). However, the high energy consumption of conventional AI architectures poses a major challenge, especially in power-constrained or remote industrial environments. This paper explores the design and implementation of low-power AI architectures aimed at enabling energy-efficient industrial automation. We examine current trends in edge computing, TinyML, and neuromorphic systems, and present a comprehensive analysis of both hardware and software-based approaches for reducing power consumption. Case studies of real-world applications demonstrate the feasibility and benefits of adopting low-power AI solutions in industrial settings. Finally, we discuss the challenges, trade-offs, and future directions for deploying scalable and sustainable AI-driven automation systems.

  • References

    [1] Lane, N. D., & Warden, P. (2018). The deep (learning) transformation of the Internet of Things. IEEE Computer, 51(9), 12–16.

    [2] Sze, V., Chen, Y. H., Yang, T. J., & Emer, J. S. (2017). Efficient processing of deep neural networks: A tutorial and survey. Proceedings of the IEEE, 105(12), 2295–2329.

    [3] Reddi, V. J., Cheng, C. H., Kandasamy, K., et al. (2020). MLPerf Inference Benchmark. arXiv:1911.02549.

    [4] Howard, A. G., et al. (2017). MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. arXiv:1704.04861.

    [5] Han, S., Mao, H., & Dally, W. J. (2016). Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding. ICLR.

    [6] Esser, S. K., et al. (2016). Convolutional networks for fast, energy-efficient neuromorphic computing. Proceedings of the National Academy of Sciences, 113(41), 11441–11446.

    [7] Zhang, X., et al. (2018). Shufflenet: An extremely efficient convolutional neural network for mobile devices. CVPR.

    [8] Nurvitadhi, E., et al. (2017). Can FPGAs beat GPUs in accelerating next-generation deep neural networks? Proceedings of the ACM/SIGDA FPGA.

    [9] Davies, M., et al. (2018). Loihi: A neuromorphic manycore processor with on-chip learning. IEEE Micro, 38(1), 82–99.

    [10] Warden, P., & Situnayake, D. (2019). TinyML: Machine learning with TensorFlow Lite on Arduino and ultra-low-power microcontrollers. O’Reilly Media.

    [11] Burhan, M., et al. (2020). IoT elements, layered architectures and security issues: A comprehensive survey. Sensors, 20(20), 5881.

    [12] Mittal, S. (2019). A survey on optimized implementation of deep learning models on the NVIDIA Jetson platform. Journal of Systems Architecture, 97, 428–442.

  • Downloads