The world’s first general-purpose humanoid robot
Figure, a robotics startup based in the United States, has unveiled its plans for ‘the world’s first commercially viable general purpose humanoid robot,’ named Figure 01.
Whilst the robot is still in its conceptual phase, the startup promises the robot will “have the ability to think, learn, and interact with its environment and is designed for initial deployment into the workforce to address labour shortages and over time lead the way in eliminating the need for unsafe and undesirable jobs.” If these plans do come to fruition, then Figure 01 could be a pioneering robot that revolutionises how we operate as a species entirely.
However, it is wise to remain sceptical about claims such as these, it is not the first time, nor the last time, that we have seen a startup emerge out of the blue with bold claims, a flashy rendering, and little proof of method. However, what makes these claims stand out is the surprisingly promising team of robotics engineers behind the project.
First up is Figure’s Chief Technology Officer, Jerry Pratt, who has a rich history in the world of robotics prior to his new role at the company. Pratt, who spent around 20 years at the Florida Institute for Human and Machine Cognition (IHMC), led the team that claimed second place at the DARPA Robotics Challenge Finals. Beyond just this feat, Pratt and his team have also worked with DRC Atlas, NASA’s Valkyrie, and Nadia, enhancing their already rich portfolio.
Founder of the company, Brett Adcock, who also founded Archer Aviation, which has successfully developed and deployed into testing a commercial passenger eVTOL aircraft, is up next, who offers an innovative mind behind the wheel of the company.
Beyond these two highlights, over the past year, Figure has been busy hiring, sniping, and persuading more than 40 engineers from a variety of sources including IHMC, Tesla, Boston Dynamics, Google X, and Waymo – most of which have prior experience specifically in humanoid robotics or similar autonomous systems.
Adcock is adamant that his team will accomplish what they have claimed is possible: “It’s our view that this is the best humanoid robotics team out there.
“We’ll have expertise in just about every part of the thousands of things that you need to do for humanoids.”
Pratt made clear that the company isn’t going to be making use of fancy new technology initially and that their designs aren’t based on some farfetched secret technology in development. Rather the team at Figure will utilise existing and future technology as it comes. “It’ll be a new design, with really solid engineering.” Says Pratt.
Currently, the team has a CAD model of the actual robot they plan to create, this is what the renders of Figure 01 show. The final product would be entirely electrically powered, reaching 1.6 metres tall, weighing in at 60kg, with a 20kg payload, and 5 hours of battery life per charge.
As Pratt explains, “having a humanoid form – it’s really tough doing the packaging. In general, with technology that’s available today, you can hit somewhere around 50 and 60% on most human specs, like degrees of freedom, peak speeds, and torques, things like that. It won’t be superhuman; we’ll be focusing on real-world applications and not trying to push the limits of pure performance.”
This is the philosophy behind the design of the robot, creating one for commercial utility instead of flashiness. A realistic humanoid robot makes sense in a world designed for humans, with this pursuit in mind you won’t get a robot that can do fancy tricks, but you will get a robot that can prove productive in cramped workspaces or navigate a warehouse safely.
The opinion of Figure is simple, for them, building a robot without legs that has the necessary range of motion that would enable it to work effectively in human workspaces would have the same level of complexity as developing one with legs in the first place. This is the justification behind the added complexity of designing a humanoid-legged robot. The added bonus of overcoming these additional challenges will also leave the final product in a much better position to generalise. Whilst Pratt does admit that “for just one application, there’ll probably always be a dedicated robot that’ll be better,” the point of Figure 01 is to be a robot that can be applied as a jack of all trades.