Researchers have built an “actually useful” working robot based on a design dreamt up by AI.
Teams at TU Delft and Swiss technical university EPFL tasked artificial intelligence tool ChatGPT to help with food production.
They followed all of ChatGPT’s design decisions, which included which crop to harvest and how to effectively pick it.
The result is a tomato-harvester that knows to gently pluck the produce from the vine.
The researchers wanted to explore the varying degrees of cooperation between humans and Large Language Models (LLM), of which ChatGPT is one.
In research published in Nature Machine Intelligence, team members Cosimo Della Santina, assistant professor, and PhD student Francesco Stella, both from TU Delft, and Josie Hughes from EPFL, asked “What are the greatest future challenges for humanity?”
Della Santina said: “We wanted ChatGPT to design not just a robot, but one that is actually useful.”
In the end, they chose the food supply as their challenge, and as they chatted with ChatGPT, they came up with the idea of creating a tomato-harvesting robot.
The researchers followed all AI design decisions, input they say proved particularly valuable in the conceptual phase.
Della Santina said: “ChatGPT extends the designer’s knowledge to other areas of expertise. For example, the chat robot taught us which crop would be most economically valuable to automate.”
ChatGPT also came up with useful suggestions during the implementation phase, such as “Make the gripper out of silicone or rubber to avoid crushing tomatoes” and “a Dynamixel motor is the best way to drive the robot”.
The result of this partnership between humans and AI is a robotic arm that can harvest tomatoes.
“However, we did find that our role as engineers shifted towards performing more technical tasks,” says Stella.
Della Santina, Stella and Hughes said they will continue to use the tomato-harvesting robot in their research on robotics. They are also continuing their study of LLMs to design new robots. Specifically, they are looking at the autonomy of AIs in designing their own bodies
“Ultimately an open question for the future of our field is how LLMs can be used to assist robot developers without limiting the creativity and innovation needed for robotics to rise to the challenges of the 21st century,” Stella said.
Produced in association with SWNS Talker
Edited by Saba Fatima and Newsdesk Manager