Researchers from Delft University of Technology and the Swiss Federal Institute of Technology (EPFL) have collaborated with ChatGPT-3 to design and construct a robot that can pick tomatoes. The researchers used the large language model to access global data from various sources to identify the key areas in which robots can help humanity. ChatGPT-3 defined three areas of focus: food supply, an aging population, and climate change.
Leveraging the LLM’s ability to process and comprehend vast amounts of data, the researchers sought advice on the features a robot harvester should possess. ChatGPT-3 provided technical recommendations, such as a motor-driven gripper for effective tomato harvesting. The research team refined the specific features, including construction materials and computer code to control the robot, while ensuring ethical and practical concerns of the work.
The researchers successfully built a robot that could pick tomatoes, thanks to the assistance provided by ChatGPT-3. This exciting collaboration showcases the potential for human-LLM collaboration in the design process. The researchers caution against making decisions based solely on AI recommendations and emphasize the importance of human expertise in managing ethical and sustainable advancements.
This collaboration demonstrates the potential of LLMs in opening up exciting possibilities for designing robots that serve humanity. With the proper management of human-AI collaborations, LLMs can be a force for good, driving the progress of robots in an ethical and empowering manner. Ethics, practicality, and efficacy are the key focuses of this innovative project.
