Robot "GEAIR" Unveiled, Paving the Way for a New Paradigm in Intelligent Breeding

On August 11, researchers led by Dr. Xu Cao from the Institute of Genetics and Developmental Biology at the Chinese Academy of Sciences published a groundbreaking study in the journal Cell, titled "Engineering crop flower morphology facilitates robotization of cross-pollination and speed breeding." The research marks a major leap in agricultural innovation by integrating biotechnology (BT) with artificial intelligence (AI) to pioneer a new paradigm of crop-robot co-design, described as a "mutual pursuit" between biology and robotics. The team developed a novel system that re-engineers plant flower structures through gene editing to create "robot-friendly" male-sterile lines—specifically, structural male sterility that allows the stigma to naturally protrude without requiring manual intervention. This breakthrough addresses a long-standing bottleneck in hybrid breeding: the high cost and inefficiency of manual pollination, especially in crops with closed flowers like tomato and soybean. In tomatoes, the majority of commercial varieties are hybrids, yet their closed flower morphology has made manual cross-pollination unavoidable for decades. Labor costs for this process account for over 25% of total breeding expenses, with manual emasculation alone contributing to 40% of those costs. With aging populations, labor shortages are driving up these costs further. Similarly, soybeans and other crops with similar floral structures have remained unable to fully harness hybrid vigor due to prohibitive manual labor requirements. To overcome this, Xu Cao’s team used CRISPR gene editing to target the GLO2 gene—a key MADS-box transcription factor in the ABC model of floral organ development—specifically disrupting stamen development. This resulted in open, split stamens and naturally exposed stigmas, enabling self-pollination prevention and efficient hybridization without manual labor. Crucially, the modified plants maintained normal fruit yield and seed quality, and the approach is broadly applicable across species, independent of genetic background. In collaboration with Dr. Yang Minghao’s team from the Institute of Automation at the Chinese Academy of Sciences, the researchers developed the world’s first intelligent robot capable of autonomous navigation and automatic cross-pollination: the "Jill" robot, or GEAIR (Genome Editing combined with AI-based Robotics). In commercial greenhouses, Jill achieved a stigma recognition accuracy of 85.1%, with an average pollination time of just 15 seconds per flower. One full robotic cruise achieved a successful pollination rate of 77.6% ± 9.4%, and the robot can operate continuously, ensuring high pollination success across multiple cycles. Notably, over 95% of the robot’s components are domestically produced in China, highlighting its strong potential for large-scale deployment and technological independence. The team further integrated Jill with two other key technologies developed by Xu’s group: the "de novo domestication" method (established in 2018) and "speed breeding" techniques. Together, these form a fully automated smart breeding factory. This system has reduced the breeding cycle for wild relatives of crops from five years to just one year, dramatically lowering labor demands and unlocking the potential of wild species to enhance cultivated crops—enabling the rapid development of flavorful, high-quality tomatoes and stress-resistant, high-yielding new varieties. The application of the Jill system to soybeans represents another milestone: the first successful creation of a structural male-sterile soybean line using gene editing. This innovation reduces manual pollination time by 76.2%, offering a transformative solution to the long-standing challenge of soybean hybrid breeding and paving the way for significant yield increases. This research establishes a new model for smart breeding—“BT foundation + AI empowerment + robotic labor” (BAR)—that redefines the future of agricultural biotechnology. It exemplifies the power of "AI for Science" in driving innovation and generating new productive forces. The work was supported by major national programs including the Ministry of Agriculture and Rural Affairs, the Chinese Academy of Sciences Strategic Priority Research Program, the National Natural Science Foundation of China, and the Beijing Innovation Team Project for Intelligent Greenhouse Vegetable Production.