|
|
|
![]() |
|
|||||||
|
||||||||
![]() |
|
|
Thread Tools | Rate Thread | Display Modes |
|
|
|
#1
|
|||||
|
|||||
|
Quote:
Last edited by Madison : 16-07-2003 at 00:38. |
|
#2
|
|||||
|
|||||
|
Quote:
|
|
#3
|
|||||
|
|||||
|
Quote:
-dave |
|
#4
|
|||||
|
|||||
|
Quote:
Mission: Space's pods, on the other, need only spin around in a big giant circle or otherwise rotate on, to my knowledge, a single other axis. It may be two axes, though. It's not easy, I'm sure, but it does seem less complex than Test Track. |
|
#5
|
|||
|
|||
|
ahh.. fond memories of the Test Track... back in 2000.. finally experienced it for the first time.. as a result.. a shut down in the 'cold test' room.. hehe.. I pittied the car that was ahead of us in the 'heat test' room
![]() |
|
#6
|
|||||
|
|||||
|
Quote:
). It is basically a big, grown-up version of the "railroad signaling system" controls problem that sophomore Computer Science students used to solve. But what I don't know are the details of the specific implementation (vs. the general algorithmic approach) that Disney used to cure the 45-mph collisions they were experiencing at one time.Don't confuse the complexity of the supporting software system with the complexity of the kinematics of the ride pods/cars. The cars on Test Track basically have just one degree of freedom each - their position along the track (with the free variable being their acceleration along the path at any moment in time). Their path in 3-space is completely pre-defined and physically restricted. The pods in Mission: Space have three degrees of freedom each (the centrifuge rotation plus 2 minor axes of motion). At the highest level, their path is pre-defined (i.e. a really big circle). But at the detailed level, the position and minor axis orientation along that circular path is variable, based on the outputs of the control system. The complexity comes in when you start adding multiple units to the system, and fully load the rides with passengers. At full load, 27 cars on the track gives 27 total degrees of freedom that have to be controlled by the Test Track software. For Mission: Space, 4 centrifuges with 1 controlled motion (rotation) and 10 pods/centrifuge with 2 controlled motions/pod gives 84 degrees of freedom that have to be simultaneously controlled. Control of any one individual motion on either ride may be a relatively simple matter, but handling the large number of simultaneous operations (and their potential for interaction), is the hard part. This is an over-simplification (for example, the Test Track cars have independent control inputs for brakes, acceleration, etc., and there are lots of controlled motions in Mission: Space associated with safety systems that aren't mentioned here), but you get the idea. -dave |
|
#7
|
|||||
|
|||||
|
I guess I assumed, and probably erroneously so, that Mission: Space's pods should all be doing the same motions at the same time. That is, the big circle spins and the remaining pods at its circumference each tilt or roll (or yaw and roll or yaw and tilt or whatever) in sync. such that there need only be two "outputs" per centrifuge. One controls its rotation on the whole and the other controls the specific positioning of the pods -- all at once. Whereas, again, on Test Track, the individual cars are all doing individual things. It stands to reason that no two cars on that ride require the same control signals at a given time. But, if all of my reasoning is based on a bad assumption about what's going on in the Mission: Space ride system, I'm just blathering.
Of course, I could probably have better results trying to perform brain surgery than I would with trying to program a computer. My experience in each area is similar. You have the neat advantage of getting to talk to all the neat folks involved, whereas I just have to sit along the sidelines and drool jealously. I really appreciate your insight, as this is one subject that I like quite a bit more than FIRST. I know, I know -- blasphemy. But, you raised another question in your examples. . . I would've expected that each of Mission: Space's four centrifuges would operate as its own unit with no ties to the remaining three. Are they all connected to a central computer? Is that just a cost-saving measure, or is there a practical benefit in each centrifuge having the ability to talk with the others? |
|
#8
|
|||||
|
|||||
|
Your basic assumptions are not at all unreasonable. Ideally, the pods in a given centrifuge SHOULD all be doing the same thing. But the subtlety is that there is no physical/mechanical interconnection between the pods to force them to all go through the same motions, and the control loops for those minor axes motions are independent of each other. Therefore, when the ride script calls for the pods to orient to "position X," ten independent control loops have to all execute to move each pod to the commanded orientation (again, this is simplified - there are additional high-level command and safety loops on top of this, but you get the idea). While the top-level script calls for all the pods to do the same thing, each pod has the capability of independent control.
A logical next question would be "why do that and add all that complexity?" While it does appear that each pod is doing the same thing, the independent control allows each pod to dynamically adjust itself based on the loading (ie. number of passengers) so the centrifuge remains balanced. The careful load balancing is one of the reasons that the ride is so smooth, and the riders get a perception of isolated acceleration without any distracting rumbling, vibrations, or "spinning" sensations. There is one top-level control system, but it is a network of multiple computers and not just one unit. Each centrifuge operates as a discrete unit, and can be dropped out or added to the ride system without affecting the other centrifuges. But they are all controlled by the same system (hint for riders: the "Mission Control Center" next to the gravity wheel that you see while waiting in line is actually the ride control system. The side of the consoles that face away from the riders house the control panels and monitors). -dave |
|
#9
|
|||||
|
|||||
|
Thanks, Dave.
I still live in a perfect world where things do what they're told to do
![]() It makes sense, though, that you'd want each pod to have the ability to make small adjustments to its positioning during the ride. Otherwise, over time, it stands to reason that each pod may drift away from its intended zeroed position, and they may not all drift the same way. There are so many questions I want to ask, but I don't know if this is the best forum to continue doing so -- if any such forum even exists. Again, thank you. |
|
#10
|
|||||
|
|||||
|
Ok, now it's up to 31 times on M:S.
And anyway, the problem with M:S is that each centerfuge needs to be calibrated and then set incase any mechanical device isn't working 100%. That's what Disney is worried about. |
|
#11
|
|||||
|
|||||
|
http://www.msnbc.com/news/948158.asp?0ql=c8p
Also on MSNBC now, are we? =) Congrats. I was reading the article with my mom peeking over my shoulder, and when I read your quotes, I yelled out 'That guy owes me doughnuts!' So now my family thinks I'm insane. |
![]() |
| Thread Tools | |
| Display Modes | Rate This Thread |
|
|
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| pic: Dave Lavery | CD47-Bot | Extra Discussion | 5 | 21-04-2004 04:09 |
| Dave Lavery | Mike Schroeder | Rumor Mill | 117 | 24-06-2003 15:19 |
| Dave Lavery is insane!!! lol Krispy Kreme Drive Through "car" | Elgin Clock | General Forum | 16 | 11-03-2003 23:19 |
| Dave Lavery in Esquire | Wetzel | General Forum | 11 | 24-11-2002 20:40 |
| Look! It's Dave Lavery in his birthday suit! | MissInformation | Chit-Chat | 17 | 02-06-2002 00:46 |