Beefy ML - Ultra-Low Power Algorithms on Cattle

Explore the innovative application of machine learning in agriculture through this conference talk on ultra-low power algorithms for monitoring cattle. Discover how robust ML algorithms can be implemented on basic Cortex M4F processors, enabling long-term monitoring of animal behavior. Gain insights into the system design, algorithm implementation, and validation process for tracking cattle feed intake using accelerometer data. Learn about the integration of open-source components, including Zephyr OS, Zephyr Power Management, PyTorch, and CMSIS DSP, to create an efficient and sustainable solution. Understand the challenges and solutions in data acquisition, annotation, and training for this unique application of embedded machine learning in livestock management.