GEAIR 2.0 robot for crop breeding unveiled at Beijing seed conference

GEAIR 2.0, described by its developers as the world's first humanoid embodied-intelligence robot designed for crop breeding, was unveiled in Beijing at the 33rd China Beijing Seed Industry Conference, according to Beijing Daily.

The robot was developed by a team led by Xu Cao at the Institute of Genetics and Developmental Biology of the Chinese Academy of Sciences. Made entirely with domestically produced components, GEAIR 2.0 can autonomously identify flowers, navigate around obstacles, and carry out hybrid pollination around the clock.

The upgraded version replaces a flexible robotic arm with a dual-arm system, improving its ability to work around leaves and branches and adapt to changing conditions, Xu told Beijing Daily. Its stigma-recognition accuracy has increased from 85 percent to 92.8 percent.

At the launch, the robot demonstrated tomato pollination using both hands, with one holding a pollen tube and the other a pollination brush. It completed a single pollination in less than 10 seconds, and one dip of the brush into pollen supports about 50 pollinations, according to China Central Television.

The technology combines robotics with biotechnology. The GEAIR system — short for Genome Editing combined with AI-based Robotics — uses gene editing to create male-sterile plants with exposed stigmas that are easier for robots to access.

This "crop-robot co-design" strategy was detailed in a Cell paper published in 2025. The study showed that automated cross-pollination in tomatoes could achieve efficiency comparable to manual pollination, and this approach was successfully extended to soybean, yielding comparable floral traits suited for robotic pollination.

Xu said hybrid pollination remains repetitive and physically demanding work that must often be completed within a short time window. Workers may need to stay bent over plants for long periods, while robots can operate for longer after charging and patrol autonomously in hot or cold conditions and at different times of day.

By combining biotechnology and AI, the researchers hope to reduce labor demand, lower breeding costs, and improve the efficiency of hybrid breeding.