Google Developed A New Robot That Can Code Itself

Google researchers believe that natural language processing and AI will enable robots to create their code to take action against new instructions. Google is currently testing a system that allows robots to write their code, follow instructions, and complete tasks.

This is a way to streamline reprogramming policies for every new task. It can be tedious and time-consuming, and it requires domain experts.

Staff could interact with robots on smart factory floors using simple commands without needing to write complicated code.

Google researchers developed language modeling programs called Code as Policies (CaP). This code-writing AI system can generate new code for new instructions.

Google robotics scientists stated in a blog post that “Given natural languages, instructions, current language models can write not only generic code, but, as we have discovered, code capable of controlling robot actions.” Google researchers have combined large language models with everyday robots to better respond to complex and abstract human requests.

According to the Google team, CaP will enable a single system to perform complex and diverse robotic tasks without requiring task-specific training. The demonstrations illustrate how robotic arms can adapt to new instructions. For example, they move blocks in a square and make them larger.

The blog post explains, “our experiments show that outputting code leads to improved generalizations and task performance over directly learning robotic tasks and outputting natural-language actions.”

They also noted that CaP provides a level of generalization thanks to the pre-trained language model, which is much more efficient than the massive amount of data required for robot learning.

The tech giant released the code on GitHub to allow others to experiment with the system.

Google researchers stated that by identifying the types of generalization encountered in code generation issues, they could also study how hierarchical code generation improves generalization.

Related Popular Stories:

Help Someone By Sharing This Article