Google has created a new large-scale learning model that improves the robot’s overall performance and ability to perform more complex and abstract tasks, as well as handle complex human requests.
The Google-Everyday Robots research, dubbed ‘PaLM-SayCan,’ employs PaLM — or Pathways Language Model — in a robot learning model.
“This effort is the first implementation that uses a large-scale language model to plan for a real robot. It not only makes it possible for people to communicate with helper robots via text or speech, but also improves the robot’s overall performance,” the tech giant said in a blog post.
Today, most robots exist in industrial settings and are painstakingly programmed for specific tasks.
As a result, they are unable to adapt to the unpredictability of the real world.
“That’s why Google Research and Everyday Robots are working together to combine the best of language models with robot learning,” said Vincent Vanhoucke, Head of Robotics at Google Research.
The new learning model allows the robot to comprehend how we communicate, allowing for more natural interaction.
“PaLM can help the robotic system process more complex, open-ended prompts and respond to them in ways that are reasonable and sensible,” Vanhoucke added.
When the system was integrated with PaLM, the researchers saw a 14% improvement in planning success rate, or the ability to map a viable approach to a task, when compared to a less powerful baseline model.
“We also saw a 13 per cent improvement on the execution success rate, or ability to successfully carry out a task. This is half the number of planning mistakes made by the baseline method,” informed Vanhoucke.
The greatest improvement, at 26%, is in planning long-term tasks, or those involving eight or more steps.
“With PaLM, we’re seeing new capabilities emerge in the language domain such as reasoning via chain of thought prompting. This allows us to see and improve how the model interprets the task,” said Google.
For the time being, these robots are simply getting better at grabbing snacks for Google employees in the company’s micro-kitchens.