ChatGPT is being used by researchers to control robots and drones.

Microsoft researchers have recently been exploring the use of language models like ChatGPT to enable natural language communication between humans and robots. ChatGPT, a large language model developed by OpenAI, has been used to train a machine learning model that can interpret human commands and control robotic movements in real time.

The researchers believe that this technology has the potential to revolutionize the way we interact with robots, making it easier and more intuitive for humans to give instructions to machines. In a blog post, the researchers explained that “using natural language as a way of controlling robots can remove the barriers between people and machines, enabling more people to access and use these technologies.”

The team has been working on a project called “CARE” (Conversational Artificial Intelligence for Robot Control in Real-Time Environments), which aims to develop a system that can interpret human language and translate it into robotic movements in real time. The system uses ChatGPT to understand the user’s commands and then generates a corresponding sequence of robot movements.

To test the system, the researchers used a robotic arm to perform a series of tasks, such as picking up objects and moving them to different locations. The system was able to successfully interpret and execute a variety of commands, including complex instructions that involved multiple steps.

The researchers believe that this technology has a wide range of potential applications, from manufacturing and logistics to healthcare and education. For example, it could be used to control robotic assistants that help people with disabilities perform daily tasks, or to enable students to learn robotics in a more intuitive and interactive way.

Overall, the use of language models like ChatGPT to enable natural language communication between humans and robots represents an exciting new frontier in robotics research. By breaking down the barriers between people and machines, this technology has the potential to make robotics more accessible and intuitive for everyone.

Via The Impactlab