4 ms·
Going from 2 to 4 legs doubles the amount of actuators required and substantially increases power consumption since you must move more mass, going to 6 compound
by solid_fuel 3mo ago
Going from 2 to 4 legs doubles the amount of actuators required and substantially increases power consumption since you must move more mass, going to 6 compounds the problem further. In a future where we have more dense power storage and better (and cheaper!) motors, you probably will see robots with more legs. But for now, the most efficient solutions are bipedal.
Especially because this thing is already $8k, I imagine they have already done some substantial price optimization.
- bensyverson 3mo agoReal question: what about 3 legs? Is tripedal locomotion a viable compromise?
- levocardia 3mo agoReal answer: how many animals can you think of with three legs?
- bensyverson 3mo agoAh, so it’s impossible to make a robot unless it looks like an animal
- scotty79 3mo agoKangooroos?
- pphysch 3mo agoThere are millions or billions of quadrupeds surviving on 3 legs every day. 4 legs is evolutionarily superior to 3 but 3 gets the job done, especially if you don't have to worry about agility or repair.
- Y_Y 3mo agoThis one right here
- wat10000 3mo agoHow many animals have wheels? And yet wheels are a pretty practical choice for a lot of things.
- scotty79 3mo agoIf the robot has two legs with wheels at the ends it could combine speed and flexibility. Adding a third leg for additional stability when the robot needs it might be great. But the third leg would probably need to be between the first two and I don't think humans are mature enough to have that.
- stubish 3mo agoThere is likely no benefit over 2 legs if you need to step over things. And if you don't, wheels are just fine. Maybe stairs changes this.
- solid_fuel 3mo agoIt's been done, I don't really know how the efficiency compares but I'm sure there's some research out there on it. The main issue I see though is that the advantage you get with 4+ legs is stability at slow speed - you can be stable on three legs while moving one, which makes precise movement easier. With three legs, as soon as you pick one up you will start to fall over, so you either need legs with enough freedom of movement to shift the center of mass of the robot back to offset the lack of support, or some other way to shift the center of mass. You have to balance with two legs, as well, but there isn't a transition from "stable" to "balancing" with every step - you're always actively balancing - which makes movement easier to plan. Overall I suspect that tripedal locomotion isn't really any more efficient than bipedal movement, and it might even be less stable. Just one example: https://spectrum.ieee.org/martian-inspired-tripod-walking-robot-generates-its-own-gaits https://spectrum.ieee.org/martian-inspired-tripod-walking-ro...
- foxylad 3mo agoConversely one leg would be cheaper and lighter. I look forward to welcoming our hopping robot overlords!
- solid_fuel 3mo agoIt's been done! https://spectrum.ieee.org/jumping-robot-salto1p https://spectrum.ieee.org/jumping-robot-salto1p
- goobatrooba 3mo agoI agree on the rest but don't see why it would be heavier. If you have more feet standing at any particular moment the load would be more distributed, i.e. movement should be easier with more legs (depending on leg weight of course).
- solid_fuel 3mo agoYou're confusing (pressure on the ground / surface area) with overall weight (the sum of all pressure on the ground). Having more legs is heavier simply because now you have additional struts and motors that you didn't before, those items are linear with the number of legs, and the total weight is weight_of_leg x num_legs.