Wow. I'm wondering if once this robot achieves sentience, will he regard this experience of stumbling through the forest much like a human regards being in the womb, who is then, upon birth, confusedly, forcibly ejected into bright light, staggering, bloody, crying at the sudden onslaught of sounds and senses that he doesn't have the capacity to make sense of?
I still don't know why they don't use gyroscopes for stabilization. It seems counterintuitive to purely focus on kinematics when they can use gyroscopes.
OR why they don't apply that evolutive walking algorithm on it.
Yeah, meant flywheels. And what do you mean, they impose inertia that isn't useful? I'm curious.
So a flywheel will keep you from falling. It will do this, however, by keeping you from moving. The physics are complicated (especially for me, considering Dynamics was like 18 years ago) but... Here, look. Ever seen a powerball?
Flywheels work through rotational inertia. Force applied against the axis of a flywheel is translated through rotational inertia in another direction. This makes them stable, and that stability is why flywheels work for positioning. BUT the more stable that flywheel, the more rotational inertia they have... which effectively means the more resistance to moving they have, not just resistance to "tipping."
Flywheels work great to keep you in the position you want to be in. If you want to be in another position, though, flywheels work against you. So while a flywheel will keep your walking automaton upright, it will also increase the amount of force necessary to, say, go up stairs. Or move to one side quickly. Or translate in a desired direction. They aren't without their drawbacks. This is one reason why racing motorcycles have carbon fiber wheels - the flywheel effect from aluminum makes them handle with less agility.
On the other hand, if your MEMS sensors can sense the way a flywheel gyro can without imparting the moment of rotational inertia you get from a flywheel gyro, you can get the benefits of a gyro without any of the drawbacks. That's why everyone went solid-state long ago.
First - I am aware of all that, though I fail to understand how rotational momentum works against translational movement. It seems EXTREMELY counterintuitive. As far as I know, gyroscopes only stop rotational movement in the plane they're not spinning in (which, in my point of view, is absolutely perfect for a bipedal considering the position of the hips won't vary much more than five degrees - unless you want to design a crawling robot, or something, it is not really an issue). Place the flywheel "horizontally" and the robot can spin around itself without falling. As for motorcycles, it is a completely different example - flywheels, in that case, are not adequate because a motorcycle's agility depends on two planes of rotation - yaw and roll. But seriously though - if you have proof that gyroscopes stop translational movement in it's frame of reference, let me know - because as far as I know, the ISS's not showing any signs of falling down, and it's not geosynchronous.
And as @acyclicks mentioned, if your robot needs to be angled at more angles than would be allowed otherwise, you can mount it on a rig with actuators or within a sphere with motors to change the angle.
Some options:
- Earth's surface and orbit are not inertial reference frames. Probably negligible effect.
- The flywheel would have mass, and would resist lateral acceleration just because of its momentum.
- Maybe some rotational movement is inherent to bipedal locomotion. You mention a 5 degree rotation in the hips which might need to act against the flywheel.
- "Dynamics was 18 years ago"
- I never took dynamics and am missing something
My question is: WHY would rotational momentum resist lateral acceleration? As far as I know, an object in motion stays in motion unless disturbed by a net unbalanced force. That implies that the flywheel would be able to apply lateral force against the movement of the robot - and if it does, how does it apply that force?
As a matter of fact - if that were true, surely there would be implications for airplanes (which have large spinning disks as part of their engine), cars (which have a flywheel in the engine), no?
For your specific points:
-If Earth's surface and orbit are not inertial reference frames, what reference frame would cause the robot to "have trouble going upstairs" because it has a flywheel?
-Yes, they flywheel would have mass. However - would the fact that it's spinning change it's translational inertia?
-Rotational IS a part of bipedal locomotion - in one plane. With a single flywheel it is not an issue. So unless the robot is planning to do Olympic-style tilt-starts and tilt-turns when running, that flywheel should not hinder movement as long as it's parallel to the ground.
Maybe I have a more plausible explanation:
You have a humanoid robot with a flywheel parallel to the ground for a head. You want him to move forward. To move forward, force is applied by the robot's feet to the ground. This is not happening at the robot's center of mass, so torque is applied to the system. This torque needs to be matched by the flywheel. Maybe that dynamic causes resistance of lateral movement.
Airplanes and cars would be mostly exempt because their flywheels don't have much angular momentum, and their torque is overcome in other ways. Airplanes would be doubly exempt because their lateral force is applied near their center of mass.
I don't really know anything, and that might have been totally wrong.
On the point-by-point:
- Earth's surface and orbit are not inertial frames due to the rotation. Lateral motion on earth's surface also implies rotation, as it is a sphere. If the robot wanted to travel 12k miles in any direction, it would need to overcome a half rotation of its flywheel. Also it would constantly want to fall down as the earth rotated.
- I just meant the inertia from the flywheel's mass alone. Unless my model above is correct, I agree the mass spinning shouldn't make any difference.
- I agree pivoting about the vertical axis is the most obviously important rotation for bipedal motion. But I think the tilt-starts and tilt-turns might be an important part of bipedal motion as well. Since all the transational force is originating at the feet, you need some way to overcome that torque. You can do it by tilting the center of mass in to the direction of travel, or try to compensate with a flywheel, or maybe attach a propeller to the robot's nose.
One question. Why put the flywheel close to the head? (Also I don't think flywheels work that way - most likely it would just look strange, having an ultrastable head - maybe, if built as to make this, have a slouched walk as nothing above the hips except the head is active, resulting in a two-point stabilization - the legs and hips, from the ground, and the gyro head.
Cars have a (by my math) 4kg flywheel spinning at, usually, 2000 to 6000 RPM. Planes have (by my estimates - I don't know the exact materials used or exact volume of moving parts, but for a JT9D - the engine used in a lot of 747's, I'm estimating about 25% of moving mass, and the engine weighs 3905 kg - so that alone would be close to 1000kg, so I'm generalizing) upwards of 500 kg of moving parts spinning at upwards of 10 000 RPM. That's a lot of energy, and far from insignificant. How insignificant? The ISS, a space building, uses only a few flywheel systems no larger than a man.
-If Earth's surface and orbit are not inertial frames, then what would be the reference frame in which one would have trouble carrying a moving flywheel upwards? Also, 12kmiles - that's about half the circumference of Earth. Let's assume, for pure terror factor, that this machine can run at about 30 mph - faster than the fastest human sprint speed recorded - constantly. To run 12k miles, it would take it about 400 hours. That's a correction of about half a degree an hour - I'm pretty sure that unless you had a supermassive flywheel, the body would have no trouble applying the force necessary to do it. (And if it didn't - we can use gyroscopic precession to apply force to the flywheel to force it to tilt). It most certainly would not fall over, though.
-The flywheel shouldn't be more than 10% (if that) of the mass of the machine. It's not especially significant. We humans deal with worse regularly.
-Walking and running doesn't create as much torque as you believe. We don't lean forwards just to walk - and we're not doing it that significantly while running. A flywheel would remove the need for that. Tilt-turns maybe - but as previously mentioned, you can apply force to the flywheel directly to make it tilt OR you can give the legs a wide angle of operation and the issue goes away.
> Why put the flywheel close to the head?
Just thought it was good imagery. It didn't really matter if it's at the hips or head for that model.
> 4kg flywheel spinning at, usually, 2000 to 6000 RPM
Assuming a 12 inch flywheel and 6000 RPM, that's roughly equivalent to 4kg moving at 200MPH. Roughly 2% of the kinetic energy of a subcompact moving at 25mph. So I mean, it's not incredibly significant.
Anyways, we all took the same physics classes as you. Nobody is arguing with you that rotational energy doesn't effect linear motion in an inertial reference frame in a frictionless vacuum.
> We don't lean forwards just to walk
We definitely shift our center of mass in the direction of acceleration to accelerate. You don't have to call it leaning, and you could implement a robot whose torso remains upright while shifting its center of mass forwards, but you have to agree with me on this one.
I feel like we're having different arguments here. I'm trying to come up with some plausible model of what klienbl00 could have meant when he said a flywheel could resist linear motion. Obviously our high school physics knowledge makes that sound implausible, but I was having fun trying to come up with an explanation other than "no you're wrong".
I was frustrated by statements like "the body would have no trouble applying the force necessary" and "doesn't create as much torque as you believe", because I was shooting for an explanation of any possible resistance, not just sufficient resistance to make a flywheel-balanced bipedal robot infeasible.
Sorry about that - I shouldn't have phrased it that way (as it IS an interesting thought experiment).
Sorry if I come across confrontational - all of the above was real curiosity about phenomenons I may not be aware of, and the phrasing itself is just how it came (I'm trying to fix this, but it's a hard progress).
But yeah, real sorry if that's how I came across.
I wonder if Killerhurtz could be talking about a gimbal mounted flywheel system normally allowed to rotate freely, but engaged to provide balance.
I can't find any examples of flywheels being used that way, so that must be impractical as well for some reason. Maybe the space/weight constraints of needing a flywheel mounted in a sphere per axis, or maybe this scheme would impart some angular momentum on the robot during stabilization?
Only reason I can see them not being used is that we haven't had balance-critical applications like these before.
Oh, except the ISS, which does use flywheels to stay at the same angle.
Power might be another issue - but it's an issue they'll have to address anyway if they want an untethered unit.
But sticking a flywheel at the centre of of gravity - in the hips - would add balance without, mostly, compromising any form of agility except maybe the ability to crawl or anything else that would require the hips to not be horizontal.
That's cool and all, but I'm kind of confused as to why they would try to go this route. I think their quadruped designs are more practical. They're just as mobile, can probably carry more weight, and most importantly, are probably physically more stable.
I think the idea is that you solve a lot of other engineering problems when you solve automated bipedal locomotion. Plus sci-fi has been hyping the (probably not super effective) robot soldier for a long time. People want to see something that looks like this.
Historically speaking, people want to see anything but that. Honda discovered people stopped calling them SkyNet when they started making the ASIMO line smaller than people.
That's an evolutionary line of increased terror, followed by abrupt cutification.
Same same.
Pretend you're a Pakistani NOT terrorist. You're just trying to get through your day. There are drones overhead that can blow you up at any minute (which is extralegal as fuck, by the way) and your primary concerns are access to clean water and earning enough from your blacksmith shop to keep your family in lentils. Aside from the aforementioned "death from above" issues and the fact that the local warlords have started flashing ISIS flags.
Suddenly, the US Army shows up as a "peacekeeping" force. Which are you more likely to think is more interested in "keeping the peace", this:
Or this?
Here's the thing: The United States will never fight a conventional ground war ever again. Period. End of line. The invasion of Iraq took what? A week? While the suppression of Iraq took ten years? And how effective was it?
There's no tactical advantage in making Robocop look like Robocop. You are going to have an easier time pacifying Lahore with the Care Bears Army than you ever will with Black Widow Company.
Your comment made me think of two things:
1) two Doctor Who episodes in which new beings slowly infiltrated Earth/London either as "cute" walking lumps of human fat, or inert black cubes that people adopted as pet rocks before the cubes started to take over the world
2) The scene in Age of Ultron where Tony Stark sends a gaggle of his peacekeeping robots to try to calm a crowd, and the crowd starts throwing rocks at them.
- I think the idea is that you solve a lot of other engineering problems when you solve automated bipedal locomotion.
Yeah, I was kind of thinking that might be why they'd look into it. Though I figure the two modes of locomotion are diverse enough that there might not be as much cross over.
- Plus sci-fi has been hyping the (probably not super effective) robot soldier for a long time. People want to see something that looks like this.
I think we'll get flying cars before we get humanoid robo-soldiers. Though, for my money, I think I'd be willing to settle for this.
My guess - quadrupeds probably aren't as good at going into houses, and tunnels - they want a robot soldier who can go anywhere a man can go. Two-legs for fighting, four-legs for pack animals.
There is actually a lot of value to creating bipedal humanoid robots, but it may take a while to be realized.
Since many things are designed for us to use, having a humanoid robot may enable the robot to learn human affordances. This may make it easier for the robot to "understand" humans and optimize its behavior.
Plus he's more "backwards compatible" with the world. Our entire infrastructure is made for things that are shaped like us. If this robot got orders to move to another country on a lark, he could theoretically steal a car a drive it if he was the right shape.
Admittedly, that's more sci-fi than the actual reason, but I do love a good story.