Towards a Real-Time and Energy-Efficient Edge AI Camera Architecture in Mega Warehouse Environment
Yusuke Asai, Yuki Mori, Keisuke Higashiura, Kodai Yokoyama, Shin Katayama, Kenta Urano, Takuro Yonezawa, Nobuo Kawaguchi · 2024
In response to the exponential growth of e-commerce, our paper addresses the transformation of warehouse operations towards mega warehouses, necessitating advanced digitalization. We focus on the integration and optimization of edge AI technologies to enhance operational efficiency, accuracy, and timeliness in logistic warehouses. We propose a novel edge AI architecture tailored to the processes of warehouse operations ensuring that digitalization aligns with operational workflows. Utilizing a real-world warehouse equipped with over 60 cameras as a testbed, we demonstrate the practical application and benefits of edge computing in logistics. Our experiments have shown significant potential improvements in energy efficiency and timeliness, crucial metrics for the successful integration of edge AI technologies in warehouses.