A system of AI agents creates experimental protocols, tests them in a virtual lab, and directs real lab robots, according to a preprint from University of Toronto researchers.
A team of artificial-intelligence agents can design a laboratory of robots and instruments, test the setup in a virtual environment, and then direct the physical workspace’s operations, researchers report in a preprint posted in October to arXiv. The work has not been peer reviewed.
A scientist describes their research programme and lab set-up to the agents. The agents then create step-by-step experimental protocols for the lab’s robots and generate a virtual version of the lab to test the protocols there. After simulation, the agents can direct the robots in the real lab to perform the research.
The agents can also design an entire lab from scratch, using a virtual lab to work out the ideal placement for instruments and robots.
In a test described in the preprint, the agents wrote more than 200 commands directing a robotic arm to move between and interact with instruments in an existing lab space. The agent simulated the steps in the virtual lab to ensure that everything worked smoothly, then directed the real-life arm to carry out procedures to measure the optical properties of nanocrystals.
Evan Collins, a bioengineer at the Massachusetts Institute of Technology in Cambridge, called the concept “very cool” and said it could drastically reduce the effort required to set up automated labs. He added that integrating the cameras and sensors the agents need to direct experiments effectively could require more resources than setting up a lab manually.
Collins also said the simulations would make it easy to plan robotic movements that avoid collisions between robots and other objects in the lab. He noted that setting up an automated lab is often a time-consuming process of trial and error followed by extensive testing and troubleshooting.
Yuchi (Allan) Zhao, a roboticist at the University of Toronto in Canada and co-author of the preprint, said that because the agents conduct all the planning and testing in a simulated environment, there is no need for researchers to “babysit the system” after it’s built.
Laboratory robots are tireless and uncomplaining, but they have narrow skill sets, limited mobility and an inability to get out of each other’s way. The agent system aims to address these limitations by coordinating robotic movements and lab setup through simulation before physical deployment.
Sources
This article was drafted with AI assistance and checked against the sources above. Company claims are reported as claims. Cover image is AI-generated.
The day's checked news, by email
One short email a day with the top stories. Free, and you can unsubscribe in one click.
Or follow for the latest news:
No comments yet. Start the discussion.
Researchers found an unidentified baby North Atlantic right whale off Anticosti Island, Quebec, in September, bringing the season's calf count to 24.
A magnitude 6.1 earthquake struck off Japan's eastern coast on November 6, 2026, according to the Japan Meteorological Agency and the U.S. Geological Survey. No tsunami warning was issued.
