Humanoid Robot: Scaling Challenge

On the next many Years, humanoid robots will change the nature of the work. Or at least, it is that humanoid robotics companies have been consistently promising, making them capable of lifting Hundreds Of Millions Of Dollar Those who run in Arabs on valuations.

A lot of robots will be required to fulfill these promises. Fickling robotics hopes “HundredsIn 2025, there is a factory in its marks robot “and a factory in Oregon that is capable of creating more than 10,000 robots per year. Tesla. Planning To produce 5,000 of your optimus robot in 2025, and at least 50,000 in 2026. Figure believes “100,000 robots are the way“By 2029. And these are three of the largest companies in a fast -moving location.

There are many financial analysts to increase this message: Bank of America Global ResearchFor example, it predicts that the global humanoid robot shipment will reach 18,000 units in 2025. And Morgan Stanley Research Estimate By 2050, there may be more than 1 billion humanoid robots, which is part of the US $ 5 trillion market.

But so far, the market for humanoid robots is almost completely imaginary. Even the most successful companies in this space have deployed only a small handful of robots in carefully controlled pilot projects. And future estimates appear to be based on an extraordinary comprehensive interpretation of jobs that are able to imagine as a competent, efficient and safe humanoid robot – which is not currently present – as mother -motionalization. Can the current reality be connected to the promised scale?

What would it take to score a humanoid robot?

Physically, thousands of, or even thousands of thousands, are certainly possible in the near period. In 2023, 500,000 industrial robots were installed worldwideUnder the original belief that a humanoid is equal to four industrial weapons in terms of robot components, the existing supply chains should also be able to support the most optimistic close-term estimates for humanoid manufacturing.

But simply construction of the robot is of course the easiest part of scaling humanoids, says Meloni wiseWho served as the Chief Product Officer in Chaplata Robotics until this month. “The big problem is demanding – I don’t think anyone has received an application for humanoids that will require several thousand robots per facility.” Large deployment, intelligent, explains, the most realistic way for a robotics company to scale its business, as any new customer can take weeks or months from onboard. An alternative approach to deploy several thousand robots to do the same thing is to deploy several hundred robots that can do each 10 jobs, which seems that most of the humanoid industry is medium to long -term betting.

While most part of the humanoid robotics industry is a belief that rapid progress in AI should somehow translate into rapid progress towards multipurpose robots, it is not there clear How, when, or if it happens. “I think many people expecting what they are going to go out like this,” says wise. “But the reality of the situation is that currently AI is not strong enough to meet market requirements.”

Bringing Humanoid Robot into the market

Market requirements for humanoid robots include extremely dull, battery life, reliability and very important things. Of these, the battery life is the most straightforward – to work usefully for a robot, it cannot do most of its time charging. Next edition of agility Marks robotWhich can handle a payload of up to 16 kg, includes a heavy “backpack” that has a battery with a charging ratio of 10 to 1: the robot can run for 90 minutes, and recharge perfectly in 9 minutes. The slimmer humanoid robots of other companies must necessarily compromise to maintain their svelte form factors.

In the operation, the digit will probably spend a few minutes after running for 30 minutes. This is because a 60 -minute digit runtime is essentially a reserve, if something happens in its scope, it needs to be temporarily stopped, not an informal phenomenon in the logistics and manufacturing environment that is targeting agility. Without a 60-minute reserve, the robot will be more likely to get out of the power mid-task and will need to recharge it manually. Consider what can be seen with a minor deployment of several hundred robots weighing more than hundred kilograms. “Nobody wants to deal with it,” make wise comments.

Potential customers for humanoid robots are very concerned with downtime. During a month, a downtime of about 5 hours will appear in a factory working on 99 percent reliability. War says that any downtime that stops something like a production line can cost thousands of dollars per minute, which is why many industrial customers expect the credibility of some and 9: 99.99 percent. War says that agility has demonstrated this level of reliability in certain specific applications, but not in terms of multipurpose or general-suitable functionality.

A humanoid robot should get normal in an industrial environment Security Requirements For industrial machines. In the past, robot systems such as autonomous vehicles and drones have quickly benefited from the immature regulatory environment. But sensible says that the approach cannot work for humans, as the industry is already heavy regulated – the robot is simply considered another piece of machinery.

Are more specific Safety standard Currently, under the development for humanoid robots, the autonomy in Boston Dynamics, R&D’s Associate Director Matt Power, explains Matt Power. He notes that his company is helping to develop International Organization Manualization (ISO) Security Standard Dynamically for Balancing Leg Robot“We are very happy that the top players in the area, like agility and figure, are involved in developing a way to explain to us why we believe that the systems we are deploying are safe,” says Power.

These standard are required because the traditional security approach to power cutting may not be a good option for a dynamically balance system. By doing this, a humanoid robot will fall, possibly make the situation even worse. There is no simple solution to this problem, and the initial approach that is expected to take Boston Dynamics with it Atlas robot The robot is to keep out of situations where simply may not be the best option to close it. “We are going to start with relatively low -risk deployment, and then expand because we create confidence in our safety systems,” says Power. “I think a methodical approach is actually going to be the winner here.”

In practice, low risk means keeping the human robot away from people. But humanoids that they are banned by which jobs can safely do and where they can move safely, they have more difficulty in finding tasks that provide value.

What are humanoids answer?

Issues of demand, battery life, reliability and safety need to be resolved before the scale of humanoid robots. But this is a more fundamental question to ask whether a bipat robot is really worth troubled.

The dynamic balance with the legs will theoretically enable these robots to navigate a complicated environment like a human. Nevertheless, demo videos show these humanoid robots either as small distances on the most stable or repeated flat floors. The promise is that what we are now seeing is the first step towards human mobility. But briefly in the moderate period, there are much more reliable, efficient and cost -effective platforms that can take in these conditions: robots with weapons, but with wheels instead of legs.

Safe and reliable humanoid robots have the ability to bring revolution in the labor market at some point in the future. But the capacity is only this, and despite humanoids enthusiasm, we have to be realistic about what it would be to change the ability in reality.

These articles are “Humanoid robot scaling” in the October 2025 print issue. “