Humanoid Robots Are Lousy Co-Workers. China Wants to Be First to Change That.
Humanoid Robots Are Lousy Co-Workers. China Wants to Be First to Change That.
wsj.com
Humanoid Robots Are Lousy Co-Workers. China Wants to Be First to Change That.
wsj.com
Humanoid robots sound cool, but to me, task-specific robots are more cost-effective and easier to maintain. Humanoid robots are versatile, but as a result, they also requires much more complex engineering and production and more energy, which might not be necessary for the work they are intended to perform. For example, while a humanoid robot may be capable of performing household chores, a robot specifically designed for tasks like vacuuming or dishwashing will typically do the job more efficiently, as it is optimized for that particular task.
As you point out, versatility is a huge feature. You can build one type of robot and use it for all kinds of different applications. This also means you can reuse the same robots for different jobs as needed instead of having to get a robot for every specific task. Another advantage is that you only need one set of parts to repair all your robots. Unitree can already make humanoid robots at around 10k a pop, and I imagine price will only keep going down.
For example, while a humanoid robot may be capable of performing household chores, a robot specifically designed for tasks like vacuuming or dishwashing will typically do the job more efficiently, as it is optimized for that particular task.
The counterpoint that is usually brought up to this is that human spaces, like for instance kitchens since you brought up the example of washing dishes, are designed and optimized for humans to operate in, and so humanoid robot design may actually be the most efficient navigating and working in these environments.
Secondly there is the argument of versatility, of not needing to design different types of robots for each specific task. And thirdly, there is the efficiency of mass production, that is by only having to produce one type of robot that can fulfill many different tasks, the production can be optimized and costs minimized.
I'm not saying you're wrong, i actually personally agree with you that there is good reason to remain skeptical of the necessity of humanoid robots, and this may just be me being ignorant but so far they seem to me more like a gimmick. However i thought I would point out a few of the most common arguments that are typically made in their favor.
Being human shaped is a wild disadvantage in many contexts as well. Creating machines allows for all sorts of geometry humans simply aren’t capable of mimicking. The idea of designing something to look human for the sake of doing so seems like a non-starter. If being human shaped is optimal for a task, so be it, but it just feels like people unable to overcome a sci-if image of the future (robot people doing everything for us) that existed before contemporary industrialization and frankly had much more to do with humans acting the parts in film and theater and ease of humans relating to human appearing objects in film and other media than any imagined industrial utility of human shaped robots.