I believe there’s a big gap between flashy robot prototypes and commercially viable products.
Humanoid robots like Optimus often seem more about securing business contracts or generating publicity, while real-world applications usually require specialized, less advanced robots, like one-arm robots or Stretch.
The development of robots like Boston Dynamics’ Spot tells us how long it takes to turn prototypes into practical products, and the transition from Handle to Stretch highlights the trade-offs between advanced demos and actual commercial viability.
I think humanoid robots are more of a long-term investment for future use in factories.
Mounted to a concrete pad of you can get away with it. On rails if you can’t. If you need to move in more than one direction, use wheels, steerable or mechanum. If you need to move over strange terrain, like stairs or rubble, then you might want to look at legs. (Tracks are another option) If you can fit 4 legs, that’s usually more stable. If you can’t, you might be able to make it work with 2. So bipedal is really a last resort if all else fails for your use case.
Humanoid robots make almost zero sense if you’re talking about practical applications. Why would you make a robot bipedal, when there are more efficient (both in powerdraw and practicality) ways to allow them to move? It’s slower, more demanding of sensors and sensors, and serves no purpose except to appear more human.
Robots design for their purpose are the only thing that makes practical sense. Even if multipurpose, bipedal still doesn’t make sense on a factory floor.
I always figured bipedal was to make robots adapt to do things humans do where humans do things.
So for a factory floor, bipedal might not make sense. But what about delivering packages to houses? Navigating a home to help someone or collect something? These types of robots are designed to be general purpose.
If you own a factory or wearhouse, ideally you design them with robots in mind. Of course until you can do that something more generally purpose might be better.
Exactly. I know folks that live on the 3rd floor of an old walk-up. Stairs are the only way in & out. So if you want to replace delivery people then it would need to be a robot that can navigate narrow & windy stairs.
i would still guess a purpose built robot to navigate stairs and carry a package would be better off with something other than two legs.
Wheeled
wheeled with a fancy suspension: (yt link)
Well shit that was awesome!!
Is this real?
I don’t own one, but the previous (non wheeled ones) have been around for a while, they’re on Amazon with mixed reviews, people on Reddit seem to have been using them too.
This list price for this bigger more performant one seems pretty high though at $100k
Factories already use a ton of different robots, none of them bipedal. You use people for when the task is too fiddly for a robotic end effector, like wiring harnesses. Or sometimes things with dynamic environments like loading/unloading boxes. For that use case, Boston Dynamics actually made a wheeled robot.
Spot’s use case is retrofitting robotics into things that weren’t meant for it. But it’s rarely the right choice if you’re building something new.
With an optimum design, bipedal robots are likely more power efficient. Read up on the efficiency of the human gait to understand why.
In practice though, it’s really challenging to get an optimum design (the current state of the art is not even close), so polypeds are better from a practicality standpoint.
The future is octopeds 🐙
Harry Vanderspiegle approves.
