DISCONTINUED
The Everyday Robot Project
Teaching robots to help with everyday life
Challenge
Challenge

Creating a robot that can be taught by anyone requires tackling some of the hardest challenges in the field of robotics

Today, robots are expensive and perform highly specialized tasks. But what if a robot could be affordable and taught by just about anyone? They could help people with whatever they needed, doing tasks we haven't even dreamed up yet.

Building robots that can operate autonomously in unstructured human environments, like our homes and offices, is a complex, unsolved problem. It requires tackling and integrating some of the hardest hardware and software challenges in the field of robotics today. The Everyday Robots team was working to create a general-purpose learning robot that could operate autonomously in unstructured environments.Their mission was to build a new type of learning robot — one that can eventually learn to help everyone, every day.

A member of the Everyday Robot Project team assembling a prototype
Journey
Journey

Robots that can help in everyday environments

Today, most robots operate in environments specifically designed and structured for them. The tasks they complete are very specific, and robots need to be painstakingly coded to perform those tasks in exactly the right way, at exactly the right time. As a result, robots are incapable of adapting to the unpredictable and unstructured nature of everyday life.

Robots sorting recycling at X's headquarters

The Everyday Robot Project focused on making robots to safely operate in human environments, where things change every day, people show up unexpectedly, and obstacles appear out of nowhere. In order for robots to be useful day to day, they need to understand and make sense of the spaces where we live and work, and adapt to them as they gain experience. This requires new forms of machine intelligence.

Design
Design

Robotic systems for an unstructured world

Perception
Cameras and sophisticated machine learning models help the robots see and understand the world
Manipulation
The robot was designed to grasp, move, and interact with all kinds of everyday objects
Navigation
The robot uses data from sensors to create an understanding of what it is seeing, hearing, and where it is in the world—allowing it to safely perform useful tasks among people in everyday environments

Teaching robots new skills

The Everyday Robot project built robots to learn from human demonstration, the experiences of other robots, and from simulation in the Cloud. They also collaborated with Google AI to explore new ways of unlocking these challenges.

Today
Today

Doubling down on robotics research

After a number of years collaborating with robotics teams at Google — including bringing cutting‑edge, natural language to learning robots - The Everyday Robot’s technology was consolidated into DeepMind’s robotics efforts in January 2023. Their hardware and insights continue to support their robotics research efforts to de-risk the path towards robotics products.