SCUD - Smart Culinary Utility Device: Leveraging edge AI for battery optimization and operation cycles

Area of Science:

• Embedded Systems Engineering
• Machine Learning
• Battery Technology

Background:

• Modern lifestyles demand simpler, automated, and efficient devices.
• Existing culinary devices have high costs, labor needs, and time consumption.
• Edge Computing and IoT enable machine learning on resource-limited embedded devices.

Purpose of the Study:

• To provide an overview of Data-Driven State of Charge (SoC) estimation for Li-Ion Batteries using TinyML.
• To enable accurate machine learning model training and deployment on micro edge devices.
• To optimize processing capabilities and enhance system resilience in embedded systems.

Main Methods:

• Overview of TinyML (Embedded Machine Learning) for resource-constrained devices.
• Discussion of hardware, signals, and design files for SoC estimation.
• Inclusion of build and operational instructions.

Main Results:

• Demonstrates the feasibility of accurate State of Charge (SoC) estimation on low-power embedded devices.
• Highlights the integration of machine learning with IoT and Edge Computing for practical applications.
• Provides a comprehensive guide including hardware specifications and operational procedures.

Conclusions:

• TinyML facilitates the transition of machine learning tasks to low-end gadgets, enhancing efficiency and reliability.
• Data-driven SoC estimation is crucial for the next generation of intelligent, automated culinary devices.
• The article addresses current challenges and proposes a future roadmap for embedded machine learning applications.