The industry may be ready for anthropomorphic robots, but are the robots ready for them?

You could easily walk across the entire floor of Automat without seeing a single humanoid. By my count, there were three in total – or, rather, three units of the same non-functioning prototype. Newara was showing off its long-promised 4NE-1 robot among more traditional form factors. There was a small photo setup where you could take a selfie with the bot, and that was it.

The absence of the Agility booth at the annual Association for Advancing Automation (A3) show was notable. At last year’s event, the Oregon company had a spectacular demonstration in which a small army of digits moved cans from a tote wall to a conveyor belt a few feet away. It wasn’t a complicated demo, but the way those two-legged robots worked together was a showstopper.

Melonie Wise, Agility’s chief product officer, told me the company decided to sit out this time around, as it currently has all the orders it can manage. And that’s exactly the purpose of these trade fairs: manufacturers and logistics companies are looking for the next technological step to stay competitive.

How big a role will anthropomorphic robots play in that ecosystem is, perhaps, the biggest question on everyone’s mind right now. In the midst of the biggest robotics hype cycle I’ve seen first-hand, many are scratching their heads. After all, the notion of a “general purpose” anthropomorphic robot contradicts decades of stereotypes. The notion of a do-everything robot has been a part of science fiction for the better part of a century, but the reality is that a single-purpose system designed to do one thing well is a must.

Although there was no physical presence, the topic of anthropomorphism was discussed quite a bit at the event. So, A3 asked me to moderate a panel on the topic. I admit that I was initially reluctant to the idea of ​​an hour-long panel. After all, the panels we do at Disrupt tend to last 20 to 25 minutes. However, by the end of the talk, it was clear that we could have easily filled another hour.

This was partly due to the fact that the panel was – as one LinkedIn commenter put it – “packed.” Along with Wyze, I was joined by Boston Dynamics CTO Aaron Saunders, Apptronic CEO Jeff Cardenas, and Newera CEO David Reger. I began the panel by asking the audience how many of the attendees considered themselves skeptical of the humanoid form factor. About three-quarters of the attendees raised their hands, which is more or less what I was anticipating at this stage of the process.

As for A3, I’d say it has entered a cautiously optimistic phase. In addition to hosting a panel on the topic at Automate, the organization is holding a Humanoid Robot Forum in Memphis this October. The move is reminiscent of the 2019 launch of A3’s Autonomous Mobile Robot (AMR) Forum, which predicted explosive growth in warehouse robotics during the pandemic.

Investors are less optimistic.

“One year after we laid out our initial expectations for a global human robot [total addressable market] “With the $6 billion estimate, we raise our 2035 TAM forecast to $38 billion, which would result in a 4x increase in our shipment estimate to 1.4 million units, leading to a much faster path to profitability with a 40% reduction in bill of materials,” Goldman Sachs researcher Jacqueline Du wrote in a report published in February. “We believe our revised shipment estimate will cover 10%-15% of hazardous, risky, and auto manufacturing roles.”

However, there are plenty of reasons to be skeptical. They’re hard to deal with when you’re in the midst of hype cycles. The amount of money currently changing hands (see: Figure’s most recent $675 million raise) gives one pause in the wake of the collapse of various startups in other sectors. It also comes at a time when robotics investment has slowed after a few years.

The biggest risk at this stage is radicalism. Every piece of new technology carries this risk, but human-like robots are a lightning rod for this very thing. Just as proponents of eVTOL see this technology as fulfilling the promise of flying cars, the concept of a personal robot servant also looks within reach.

The fact that these robots look like us leads many people to believe that they can do the same things as us – or will soon be able to. Elon Musk promised a robot that would work all day at the Tesla factory and then come home to cook for you, which just added fuel to the fire. Lowering expectations isn’t really Musk’s job thingYou know? Meanwhile, others have floated the notion of a general intelligence for humanoid robots — something that seems a lot further off (“five to 10 years” is a time frame I often hear about).

“I think we need to be careful about hype cycles, because we eventually have to deliver on the promise and the potential,” Cardenas said. “We’ve been through this before with the DARPA Robotics Challenge, where there was a lot of excitement about it, and we came out of that and were hit by reality.”

One source of disconnect is the question of what these systems can deliver today. The answer is unclear, partly due to the nature of partnership announcements. Agility announced it was working with Amazon, Apptronic with Mercedes, Figure with BMW, and Sanctuary AI with Magna. But every partnership so far should be taken for what it is: a pilot. The exact number of robots deployed in any specific partnership is never disclosed, and the figure is often in the single digits. That’s exactly right: these are all operational factories/warehouses. It would be extremely disruptive to introduce a new technology at scale and hope for the best.

For this reason, pilots are important, but they should not be construed as a time for market fit. As of this writing, Agility is the only one of the group that has confirmed with TechCrunch that it is ready for the next step. On the discussion panel, Wise confirmed that Agility will announce specifics in June. Meanwhile, Cárdenas said the company plans to do heavy pilots in the “back half” of 2024, with plans to ramp up from early next year.

Talks between Newera and Boston Dynamics are still in the very early stages. Newera has promised to show off some demos sometime in July, including a series of rendered videos of the 4NE-1 beyond what has been shown so far, as well as the non-functioning units shown in Automate.

As for when we’ll see the Electric Atlas other than in a 30-second video, Saunders says, “[the video] This is just a sneak peek. We plan to get into the pilot and some more practical parts next year. As of now, we are mainly focused on designing and building out the technology. There are a lot of hard problems left to solve in the manipulation and AI space. Our team is working on it right now, and I think as these features get more robust, we will have a lot more to show.”

Boston Dynamics isn’t starting from scratch. More than a decade after Atlas, the company has as much human expertise as any other, while the launch of Spot and Stretch has taught the firm a lot about commercializing products after decades of research.

So, why did it take the company so long to move into the commercial humanoid category? “We wanted to make sure we understood where the value lay,” said Saunders. “It’s really easy to make demo videos and show cool stuff, but it takes a lot of time to get an ROI [return on investment] matters which justify the human form.”

Of the companies on stage, Neura easily has the most diverse portfolio. In fact, one gets the impression that whenever the company is finally ready to launch a humanoid, it will be just another form factor in the company’s portfolio, not the driving force. Meanwhile, when the electric Atlas finally launches, it will be Boston Dynamics’ third commercially available product.

Since Digit is Agility’s only offering for the time being, the company is fully committed to the bipedal anthropomorphic form factor. For its part, Apptronic splits the difference. The Austin-based firm is taking a best tool for the job approach to form factor. If, for example, legs aren’t needed for a specific environment, the company can mount the upper half of its robot on a wheeled base.

“I think at the end of the day, it’s about solving problems,” Cardenas said. “There are places where you don’t need bipedal robots. My view is that bipedal form factors will win the day, but the question is how do you actually get them there?”

Not every terrain requires legs. Earlier this week, Diligent Robotics co-founder and CEO Andrea Thomaz told me that her company targeted healthcare first because of the prevalence of ADA (Americans with Disabilities Act) compliant structures. Wherever a wheelchair can go, a wheeled robot should be able to follow. Because of this, the startup didn’t face the very difficult problem of building legs.

However, legs have many advantages beyond the ability to handle things like stairs. Reach is a key factor. It’s easier for legged robots to reach low shelves because they can bend from the legs and waist. Theoretically, you could add a very large arm to the top of an AMR, but doing so introduces a variety of new problems, such as balance.

Safety is something that has received less attention in conversations about form factors so far. One of the main selling points of humanoid robots is their ability to slot into existing workflows alongside other robotic or human colleagues.

But these robots are large, heavy and made of metal, so they pose a potential hazard to human workers. This topic is particularly top of mind for Wise, who says more standards are needed to ensure these robots can work safely alongside people.

For my part, I have been advocating for a more standardized approach to robot demos. The videos of humanoids in particular have left a lot of confusion about what these robots can and cannot do today. I would love to see disclosures about playback speed, editing, the use of teleops, and other tricks of the trade that can be used (intentionally or unintentionally) to deceive the audience.

“It’s very hard to identify what’s progress and what’s not,” Wise said, referring to some recent videos of Tesla’s Optimus robot. “I think one of the things we, as a community, can do better is be more transparent about the methods we’re using. That’s giving more power to the hype cycle. I think the other problem we have is that if we look at what’s happening with any of the manned robots in the field, the safety is not clear. There’s no e-stop on Optimus. There’s no e-stop on many of our robots.”