The global site of the UK's leading magazine for automation, motion engineering and power transmission
13 July, 2024

Twitter link

Google robot software firm emerges from 5 years of stealth r&d

26 July, 2021

Google’s parent company, Alphabet, has launched a business that aims to make industrial robots easier to use, less costly and more flexible. The announcement of the business, called Intrinsic, follows more than five years of stealth development inside Alphabet’s secretive X "moonshot factory" which works on future technologies.

Intrinsic is being led by Wendy Tan White, a British entrepreneur who founded the Moonfruit Web site development business in the 1990s and was awarded an MBE for her services to technology and business in 2016. She moved to California 2½ years ago to become a vice-president at X.

According to Tan White, the software tools that Intrinsic is developing will allow more people to use industrial robots to make new products, businesses and services. “Intrinsic is working to unlock the creative and economic potential of industrial robotics for millions more businesses, entrepreneurs, and developers,” she declares.

She is hoping “to support a shift towards a more sustainable and equitable way of making things”. She points out that currently ten countries make 70% of the world’s goods. Making robots easier and cheaper to use would allow more manufacturing to be done locally, cutting transport costs and emissions, and creating jobs around the world.

In a blog announcing the new company, Tan White says that “the surprisingly manual and bespoke process of teaching robots how to do things, which hasn’t changed much over the last few decades, is currently a cap on their potential to help more businesses. Specialist programmers can spend hundreds of hours hard coding robots to perform specific jobs – like welding two pieces of metal, or gluing together an electronics case. And many dexterous and delicate task – like inserting plugs or moving cords – remain unfeasible for robots because they lack the sensors or software needed to understand their physical surroundings.

“Over the last few years,” she continues, “our team has been exploring how to give industrial robots the ability to sense, learn, and automatically make adjustments as they’re completing tasks, so they work in a wider range of settings and applications. Working in collaboration with teams across Alphabet, and with our partners in real-world manufacturing settings, we’ve been testing software that uses techniques like automated perception, deep learning, reinforcement learning, motion planning, simulation, and force control.”

In one example, cited by Tan White, the Alphabet researchers trained a robot in two hours to perform a USB connection task that would have taken specialists hundreds of hours to program. They have also coordinated multiple robots to assemble furniture and architectural installations. In one project, the team worked with researchers from the ETH Zurich university in Switzerland to coordinate four ceiling-mounted robots that bring together wooden panels simultaneously to be glued together to form a pod structure.

Intrinsic researchers believe that the software tools they are developing could transform robotics in a similar way to how the arrival of the PC transformed computing

“None of this is realistic or affordable to automate today – and there are millions of other examples like this in businesses around the world,” Tan White says. “This all hints at the potential for Intrinsic’s software to radically reduce the time, cost, and complexity required to use industrial robots – and therefore their long-term potential to help with a much wider range of problems and drive up the diversity of goods that can be produced affordably and sustainably.”

The Intrinsic researchers have been exploring how techniques such as automated perception, deep learning, motion planning, force control and simulation can be combined to make industrial robots more useful and flexible.

Their tests have shown that robots can learn how to perform dextrous tasks and apply lessons learned in one task to another similar task. They believe that industrial robotics is on the cusp of a similar shift to the one that occurred when personal computers gave people and businesses access to previously unattainable computing power.

This shift is being helped by several converging trends including the falling costs of robot hardware and sensors, and advances in machine vision and learning, which allow robots to respond to their surroundings and to learn how to perform more dextrous tasks.

Now that Intrinsic has left the X rapid prototyping environment, it will concentrate on developing its products and validating its technologies. It is hiring staff in the US and Germany – including AI experts, software engineers and roboticists – and is looking for partners in various industries that are already using industrial robots.

  • To view a digital copy of the latest issue of Drives & Controls, click here.

    To visit the digital library of past issues, click here

    To subscribe to the magazine, click here



"Do you think that robots create or destroy jobs?"



Most Read Articles