A wide range of AI-related innovations were presented at the finals of the third China's innovation challenge on Artificial Intelligence Application Scene (CICAS) from Friday to Sunday in Suzhou, east China's Jiangsu Province.

The competition, jointly sponsored by the Chinese Association for Artificial Intelligence and other institutions, focused on key areas including large AI models, generative AI and embodied intelligence. It attracted more than 3,250 teams from China and abroad, with 113 advancing to the finals.

The event comprehensively tested the real-world applicability of core technologies and their industrial implementation. Through on-site dataset evaluations, functional testing and algorithmic assessments, a total of 12 teams were awarded the top prize.

He Yiming, a Ph.D candidate at the School of Mechanical Science and Engineering (MSE), Huazhong University of Science and Technology (HUST), introduced his team's project, a human-machine collaborative intelligent operation and maintenance system based on edge AI.

"Our project presented today is essentially an edge AI-based human-machine collaborative intelligent operation and maintenance system. In short, it is a health check system for industrial equipment, which can be understood as a 'robot doctor.' Traditionally, the diagnosis is performed by regular manual inspections, relying much on expert knowledge and experience. We now achieve overtaking by changing lanes with AI technology," he said.

Edge AI refers to the deployment of AI algorithms and AI models directly on local edge devices such as sensors or Internet of Things (IoT) devices, enabling real-time data processing and analysis without constant reliance on cloud infrastructure.

In addition, a catalog focusing on nationwide AI application scenario innovations was also released at a concurrent event.

The catalog systematically outlines 100 high-value application scenarios across five major areas, including intelligent manufacturing, smart healthcare and transportation, providing key support for the exploration of scenario-driven AI application.