Log in

No account? Create an account

Sat, Sep. 25th, 2004, 09:18 am

I recently had the opportunity to try out Trevor Blackwell's segway and eunicycle. They're both a lot of fun, although the eunicycle requires actual practice to ride, unlike the segway, which you basically just step on and ride around.

I think that appropriate modifications to the eunicycle could make it even easier to ride than the segway. While this is extremely counterintuitive, it makes a lot of sense from a physics standpoint. A car is completely based on static balance, meaning that if it turns off it doesn't roll over. A bicycle has some dynamic balance, because it would fall to the right or left if the rider stopped balancing. A unicycle is completely dynamic balance, which makes it much more difficult to ride, but also makes it far more controllable and maneuverable once you've learned to ride it. The way one shifts one's weight on a unicycle, and by extension a eunicycle, is ironically more intuitive for humans than the way it works on a bicycle, because it's the same as the way we do it with the form of locomotion we inexplicably use all the time, which is bipedalism. Bipedalism is completely based on dynamic balance, and requires constant corrections even when just standing still.

In order to make a eunicycle easier to ride, it must be made self-balancing. On a unicycle, and the eunicycle as it is today, you balance by waving your hands around, which is both difficult to do and very limited in the amount of force it can exert, which makes going at very high speed inherently dangerous since you can't maneuver. A more stable vehicle can be constructed as follows: On the bottom, there's a wheel like in the existing eunicycle. Above that, there's a horizontal flywheel attached to a high-torque motor which is used for balancing. Above that, and attached to the wheel via structural components which go around the flywheel, is a platform which the rider stands on, and in front of the rider there are handlebars of the same design as the segway for the rider to hold onto.

If the rider is moving forwards and leans right, the eunicycle turns the flywheel to the left, thereby turning the rider and wheel to the right and keeping the rider from falling over (although generally this will result in the vehicle being angles slightly more forward, so it will then accelerate to keep from falling forwards). Likewise, if the rider is going forward and leans left the flywheel is used to turn the rider and wheel to the left. If the rider is going backwards then the wheel is turned left to compensate for leaning right and right to compensate for leaning left.

A weird problem is that if you keep turning in the same direction for a while the flywheel might build up considerable angular momentum, eventually getting to the point where it can't turn any faster and hence the steering bottoms out. I'm not sure if this would be a real problem in practice, there are several ways the effect could be dampened or avoided if it does.

In principle this sort of vehicle should be able to go faster than a motorcycle, since it has only one wheel and hence half the friction, although in practice the weight of the flywheel and stability issues might limit its speed.

If you really wanted the ultimate high-speed vehicle, it would probably be a glycerine jet-propelled unicycle with electronic stabilization system, although that would be incapable of idling and be in some ways closer to a jet pack than a land vehicle.

Sun, Sep. 26th, 2004 02:04 am (UTC)

The Big Win of the Eunicycle over the Segwayalike is that it's half the weight, so you can carry it around easily. It would seem that adding a large flywheel might counteract that advantage...

Pleased to see the dual-posting experiment, by the way - I hope it works out!

Tue, Sep. 28th, 2004 08:08 pm (UTC)

The flywheel does indeed suck in a number of ways. It's heavy, big, requires another big motor, and has that building up speed problem.

A much simpler approach would be to attach wheel to the seat using an axle with a small motor attached to it which just kept the wheel pointed in the right direction. The rider would then have to worry about being pointed in the right direction, which they could do with the sort of arms flailing you have to do on a unicycle just to keep stable. In this case it doesn't have to be nearly as controlled and precise, but you still have to do it to reorient yourself properly. A new rider would then probably be incapable of making the vehicle fall over, but would have a lot of difficulty staying oriented in the right direction!

Tue, Sep. 28th, 2004 03:12 pm (UTC)

The flywheel is confusing, because I think it would cause the Eunicycle fall over when it spun up (spinning the cycle the opposite direction). Also, if you wanted to hop off and pick it up, you'd have to manage with it twisting about as the flywheel slowed down, or hold onto it and use some sort of brake for it.

Either way, a very interesting read.

Tue, Sep. 28th, 2004 08:00 pm (UTC)

The flywheel isn't normally spinning. Usually the flywheel is standing still, and when the vehicle needs to turn the flywheel turns then stops. Imagine that instead of having a wheel the device is mounted so it can spin on an axle embedded in the floor. If you're turned the wrong way the flywheel starts spinning, keeps spinning until you're oriented in the right direction, then comes to a screeching halt. Due to the conservation of angular momentum, you will always be not spinning when you aren't spinning relative to the flywheel. But if you were to be spinning when you got on, the flywheel would have to start spinning to keep you oriented in a single direction, and have to keep spinning in order to maintain that, again due to the conservation of angular momentum.

The potential problem is gravity. Say that instead of being mounted to the floor the device was mounted to the wall. It could still reorient you, but gravity has this annoying tendency to exert a constant force. Say you're facing upwards and put your arm out to the right. This will cause you to rotate clockwise, which the flywheel can then counteract by accelerating clockwise itself. The problem is that the flywheel has to keep accelerating as long as you hold your arm out just to keep you pointing in the same direction. Eventually it will get to the maximum speed it can go, and then fail.

Sat, Jan. 29th, 2005 03:09 am (UTC)
(Anonymous): random uni-rider commentary

"which makes going at very high speed inherently dangerous since you can't maneuver."
...actually, I think that (like on a bicycle) you're more likely to be securely balanced when going fast, because of momentum. (This isn't true on a normal unicycle, because it's direct-drive, and it's hard to pedal really really fast; but going moderately fast is much easier than going slow, or idling.) Also, I find that having learned to ride, I don't flail my arms about much at all (save when going very slowly). Lastly - the only time a flywheel exerts torque is when it's accelerating; thus, it couldn't come to a "screeching halt" without exerting torque opposite that which it had screechingly dealt, trying to steer, and undoing whatever it did. But enough of that...

I think that, as long as you have to learn some unicycle-balance to ride a motorized one, what seems most useful would be simply a uniycycle with a trottled motor (perhaps controlled by rock-pedals mounted on the fork, where you put your feet for tricks/coasting). The user still deals with all the balance issues, but the motor takes over the physical limitations (namely, pedaling ability).

It's also an interesting question what happens if the user manages to fall off. (Probably addressed by the Seguay.) Does the Euni continue to drive itself around? It seems like a pressure senser would make sense - especially so when you pick the thing up it doesn't try to wriggle itself vertical.