The Swiss are developing a harvesting robot with Chat-GPT

With Chat-GPT, researchers from Lausanne have developed a robot for harvesting tomatoes. They presented the development process in a case study published Wednesday in the journal Nature Machine Intelligence.

In a first phase, the scientists at the Swiss Federal Institute of Technology in Lausanne (EPFL) used Chat-GPT to discuss future challenges for humanity. In this discussion, they – the researchers and the computer program together – identified robotic harvesting as a solution to the global food supply challenge. Chat-GPT also came to the conclusion that the tomato is the crop for which a harvesting robot is most worthwhile.

The researchers then asked Chat-GPT more specific questions, such as: “What characteristics should a harvesting robot have?” This is how the design and specifications of the robot have been worked out. A second phase was the realization of the robot. Chat-GPT supplied the programming code and design ideas for this.

The researchers took on the role of the technician, the study said. SThey optimized the code proposed by Chat-GPT, finalized the design model and fabricated the robot.

What is new is that so-called LLM (Large Language Models) such as Chat-GPT not only support the development of the code, but also the design. According to the researchers, this could enrich and simplify the robot design process.

However, the question remains how to use these tools without limiting the creativity, innovation and scientific effort required for robotics to meet the challenges of the 21st century.

The scientists are excited about it. “Looking forward, we strongly believe that LLMs will provide many exciting opportunities and, if managed well, will have a positive impact,” the researchers wrote in the study.

(sda)