This week, Google announced that it has trained robots to sort through office trash and remove items that should go in recycling or the compost bin.
The breakthrough came from Google X’s Everyday Robot Project, an open-ended research initiative aimed at trying to integrate robots into daily life.
Over the last few months, Google’s office robots have decreased waste contamination levels from 20 percent to just five percent.
Scroll down for video
Google X has developed self-teaching robots (pictured above) that are able to separate recycling and compost from ordinary office trash
Waste contamination levels are a measure of how much improperly sorted material there is mixed into the trash.
For the trash sorting project, Google’s decided that instead of trying to write complicated instructions for how to identify different kinds of items, they would see if the robots could just figure things out through trial and error.
Each day, humans would observe the robots as they sorted trash and flagged correct and incorrect choices.
Each night, the new data generated that day would be used to autogenerate thousands of new simulations the robot’s AI program would run, according to a blog post describing the project.
The following day, results from the simulations would then be reincorporated into the software running the robots in Google’s office and the cycle would begin again, hopefully with slightly improved performance.
The company’s trash sorting robots are around five feet tall and navigate through the office on battery-powered rovers.
The Everyday Robot project at Google is focused on teaching robots how to handle ordinary tasks without needing specific instruction
The robots are learning how to water plans and identify important and unimportant objects independently
The robots are also learning how to make navigational choices that would seem automatic to a human but could be confounding to a robot
They have a set of cameras in an enclosure that operates as a head and two arms that can operate independently to pick up and hold objects.
The rover design includes an open storage tray where they can set multiple items.
Google’s vision of everyday robots are autonomous helpers that work in offices and homes without needing oversight or instruction
The robots visual sensors help it scan the environment to identify objects and obstacles
The robots color code different items in the environment and identify what might need to be moved, cleaned up, or thrown out
The robot identifies an empty coffee cup on a worker’s desk and picks it up without being asked
WHAT IS THE EVERYDAY ROBOT PROJECT?
The Everyday Robot Project is an initiative at Google X, the company’s experimental ‘moonshot’ division.
The project researchers ways to for robots to assist humans with mundane chores and everyday tasks.
Some proposed uses would be using robots to tidy up a cluttered desk, water plants, or pick up items left on the floor.
While sorting trash doesn’t seem like an immediately important task, Google’s engineers believe it can test how effectively robots can adapt to the many small ambiguities built into human spaces.
‘Everyday environments like our homes or offices aren’t governed by a set of straightforward rules that robots can follow,’ Google X’s Hans Peter Brondmo writes.
‘Think about the people around you right now, wandering as they walk and text, or stopping suddenly to have a chat or make a detour to the refrigerator.’
‘Even everyday objects, from chairs to coffee cups, appear, move, and disappear in ways that we expect and anticipate, but that are very mysterious to a robot.’
‘Where humans naturally combine seeing, understanding, navigating, and acting to move around and achieve their goals, robots typically need careful instruction and coding to do each of these things.’
Helper robots are an everyday presence at the offices of Google X, one of the company’s experimental ‘moonshot’ divisions tasked with radical long term research projects
After its initial positive findings, Google believes trash could be just the tip of the ice berg.
The company next plans to begin testing whether it can design robots that are capable of using lessons learned from one task, like sorting trash, and apply them to a completely separate task.
Brondmo cautions that real consumer-level applications are still years away, and compares the current state of robotics to where mainframe computers were in the 1960s and 1970s.