A major leap for Materials Intelligence: Prof. Haitao Zhao’s team has published their latest breakthrough, 'Agentic Robotic Boxes for Perovskite Solar Cell Fabrication with Recipe Language Model,' in the April 2026 issue of Engineering. This research pioneers the use of AI and Robotics to revolutionize how we fabricate next-generation solar cells.
Traditional materials research is often hindered by trial-and-error synthesis, where numerical datasets lack the semantic depth necessary for recipe optimization. To overcome this, we introduce a domain-specific Recipe Language Model (RLM) integrated with an innovative 'robotic box' platform, demonstrated through perovskite solar cell fabrication. This synergy enables a self-evolving AI-Robotics discovery loop, where specialized language modeling and modular hardware iteratively refine one another, marking a significant step forward in Physical AI for Materials Intelligence.
This breakthrough further cements the global research leadership on Materials Intelligence of PolyU’s Department of Electrical and Electronic Engineering and the Research Centre for Materials Intelligent Manufacturing.
This achievement is the result of an international collaboration involving world-class institutions, including the EPFL, the University of Oxford, PolyU–WIT Research Centre for Materials Intelligent Manufacturing, Wenzhou Institute of Technology (WIT), alongside specialized State Key Laboratories in Ultra-Precision Machining Technology, Alternate Electrical Power System with Renewable Energy Sources, and Artificial Microstructure and Mesoscopic Physics.
As the flagship journal of the Chinese Academy of Engineering, Engineering has emerged as a premier interdisciplinary journal in the company of Nature and Science.
The journal paper is available in Engineering at the following link:
