The robotics industry loves a good, healthy debate. Recently, one of the most intense ones centers around humanoid robots. This has been a big topic for decades, of course, but the recent proliferation of startups like 1X and Figer — along with projects from more established companies like Tesla — has brought humanoid robots back into the spotlight.
Form factor advocates point out that we built our world to fit ourselves, so it makes sense to build robots like us to fit it. There are also certain advantages in terms of accessibility, ability to climb stairs, and versatility that come with our design.
Of course, anyone who tells you that the human body is the pinnacle of organic machinery is either wrong or lying to you. I’ve spent the last year dealing with something called “degenerative disc disease,” which is perfect proof of our imperfect design.
The form factor also flies in the face of decades of conventional wisdom that has championed single-purpose robots — machines that are designed to do one thing very well more often than not. Then there’s the whole “general purpose” part, which tends to get bogged down without much thought given to its underlying complexity.
However, humanoid robots can now claim a big technical name among their ranks. Bill Gates this week List issued Of “cutting-edge robotics startups and labs that I’m excited about.” Among the names are three companies focused on developing humanoid robots. The first and most notable is Agility, whose digital robot looks the least human-like of the three. Also included are Apollo maker Apptronik and UCLA’s RoMeLa (Robotics and Mechanisms Laboratory), which is behind the ARTEMIS soccer play.
Here’s what Gates said about Uptronic,
What is more useful: multiple robots that can each do one task over and over again, or a single robot that can multitask and learn to do more? For Uptronic, an Austin-based startup that spun out of the University of Texas’ Human-Centered Robotics Lab, the answer is clear. So, they’re building “general-purpose” two-pedal humanoid robots like Apollo, which can be programmed to do a wide range of tasks, from carrying boxes in a factory to helping with household chores.
“If we want robots in our environments to work as smoothly as possible, perhaps those robots should be modeled after humans,” he notes writing about Agility. Digit currently leads the pack in terms of real-world deployments, including a recent beta in Amazon warehouses that helped pave the way for Figure’s recent BMW deal.
Other companies mentioned in the article include machine perception company Field AI and Tevel, which makes apple-picking drones.
An endorsement like that might not move the needle very much in the human direction, and Gates is not a robot scientist at all. However, it’s rewarding to see the form factor continuing to gain more mainstream legitimacy by the day.