|
|
|
![]() |
|
|||||||
|
||||||||
26-12-2010 18:03
JVNThis is very intriguing to me. I'm assuming the middle (omni) wheel is offset lower than the traction wheels. This would mean the chassis is always sitting on 2-traction wheels + 2-omni wheels at any given time.
The weird thing is... I imagine the turning point of the robot would drastically shift from the front to the back depending on which set of wheels the robot is sitting on. (At least, this is what physics tells me).
Have you driven it yet? How does it handle?
Is there a reason you didn't go with front/read omnis and middle traction wheels? This would result in the same 2+2 configuration, but you'd have a consistent (centered) turning point.
-John
26-12-2010 19:56
nighterfighterSilly John...
That IS the whole chassis!
The game this year is obviously going to be a scooter-like thing!
26-12-2010 22:12
Mr_II tend to agree with JVN's comments, with this addition: Seems to me that by putting the omni's in the middle (presumably dropped the obligatory 1/8"), you lose your resistance to sideways pushes.
26-12-2010 22:24
ttldominationAnd are we looking at the outer part of the chassis? I see two sprockets on the middle wheel and one sprocket on each outer wheel, so I'm wondering where the motor goes...
26-12-2010 22:37
akoscielski3
Is it realy strong, i would like to do something like this because our normal chassis is extremely heavy, but strong 
26-12-2010 23:21
MattC9I'm guessing the center wheel IS DROPPED so it will function as a normal 6WD chassis. This is so because we all know on most 6WD there are only 4 wheels on the ground so it can rock and turn? right? so the middle wheels act as the turning point so its like a 4WD with omni's in front.
Do i make sense? Because it was a little hard to explain trough the 'net
26-12-2010 23:31
s_forbesIf the CG is fairly balanced front to back I don't think the traction wheels would get enough grip to greatly affect how the robot turns in place. I suspect you could probably pull some of those fun maneuvers from Lunacy though, like the sideways slides and the backwards-180-forwards whip around. Looks like fun to play with!
26-12-2010 23:33
theprgramerdudeThe middle wheel will simply make it so only a front set and middle set or back set and middle set are on the ground at a time. Thus, it's treated like any four wheel drive, except the issue with this 4 wheel drive will be that half of the mass is located beyond the sliding wheel. It certainly appears that resistance to rotation from robot-to-robot contact will be reduced, but this may be intentional (It's an interesting concept that could be explored, IMO).
26-12-2010 23:41
jsasakiThis is great
I put this up to see what kind of comments i would get ..
to clarify things up... we can swap all 3 wheels and I think we were also looking at going for all 6 wheels to be plaction style wheels.
we are going to run it on friday soooo we'll see
27-12-2010 00:37
JVN|
This is great
I put this up to see what kind of comments i would get .. to clarify things up... we can swap all 3 wheels and I think we were also looking at going for all 6 wheels to be plaction style wheels. we are going to run it on friday soooo we'll see |
27-12-2010 01:47
Chris is meI think this is really cool. Imagine a robot that could shift its CG at will by moving an arm or a weight. You could pivot among whichever axis you chose, if for whatever reason you didn't want to rotate about your center.
27-12-2010 08:44
IKEI am with JVN, give it a spin (pun intended) and shoot some video. I would suspect it would have a very crazy behavior if you had a slightly rear biased CG. Take the 2008 game, where you are trying to do laps. If the weight is rear biased, going into the turn I would suspect the robot to initially push (understeer), but then as the weight shifts to the front, and those wheels grab, it would have crazy snap oversteer. With a lot of practice, you could probably make an "FRC Tokyo Drift" video...
27-12-2010 21:56
,4lex S.Interesting, I think this wheel configuration would be handy in a way. If you ended up pinned, you could shift CG and slip out of it in a lot of cases. My old team designed mechanisms to make the robot 'slippery' in '09 and it worked fantastically. If you can shift CG and need to escape pinning this could be viable.
29-12-2010 10:44
Brandon Holley
Unless theres some sort of optical illusion going on here, it seems that the omni wheels are smaller than the traction ones. The sprockets on all the wheels appear to be the same size as well, meaning the ground speed of the omni and traction wheels will be different. I still think its worth a shot at trying and showing everyone here your results, just wanted to point that out.
-Brando
29-12-2010 18:08
PAR_WIG1350|
Unless theres some sort of optical illusion going on here, it seems that the omni wheels are smaller than the traction ones. The sprockets on all the wheels appear to be the same size as well, meaning the ground speed of the omni and traction wheels will be different. I still think its worth a shot at trying and showing everyone here your results, just wanted to point that out.
-Brando |
31-12-2010 01:53
Chris is meWell that's unexpected. It looks like it's driving exactly like any other drivetrain. Is the omniwheel touching the ground at all? Is it dropped lower than the other wheels, or are all the wheels collinear?
31-12-2010 02:22
jsasaki|
Well that's unexpected. It looks like it's driving exactly like any other drivetrain. Is the omniwheel touching the ground at all? Is it dropped lower than the other wheels, or are all the wheels collinear?
|
eventually I would like to put on supershifters and run it with that.
31-12-2010 02:26
IndySam
|
Well that's unexpected. It looks like it's driving exactly like any other drivetrain. Is the omniwheel touching the ground at all? Is it dropped lower than the other wheels, or are all the wheels collinear?
|
31-12-2010 03:15
theprgramerdudeIt looks like the front wheel isn't being used at all, like it's acting just as a 4WD with omni's on the front for sliding.
31-12-2010 04:28
Chris is me|
But it's on a hard surface. Most any drive will work well on that. Carpet will tell the real tale!
|
31-12-2010 09:52
Chris FultzFrom the video, it looks like most of the weight is on the back, and the robot is acting like a 4 WD, with the omni's being the "front" wheels. The other "front" wheels do not appear to be in contact with the surface.
When it spins, it is turning around a center somewhere between the back and middle wheels - hard to tell by just doing stop frames on the video.
As others said - try it on carpet. The carpet will give a little and absorb the wheels, whick may allow all 6 to have some surface contact.
31-12-2010 10:15
PAR_WIG1350It did seem to oversteer a bit and hit the pole in the beginning, but that could be due to lag or driver error.
31-12-2010 12:44
JVNInteresting. Not quite what I expected.
My guess as to what we're seeing:
Is the CG very close to the center? This would put so much weight on the center wheels it would offset the "zero" lateral traction affects of the omni wheels and shift the CG back towards where you'd expect for a "standard" 6WD.
-John
31-12-2010 15:42
IKEVery cool. Thank you so much for the testing and videotaping. Another really cool attribute to the video was the voice explaining what the driver should do as opposed to random driving around.
As far as turning goes, I was impressed that the turing center was so closely loacted to the center of the wheels (most notable at the spin near the end of the clip). I would tend to concur with JVN that maybe most of the weight is on those center wheels.
As fat as my over-steer under-steer comments from before, it would likely take a little more speed for these behaviours to show up. The bot appeared to be traveling at or below 4 ft/s.
Published lateral grip on those wheels is supposed to be around 0.2. I would expect the oversteer understeer stuff to come up when attempting at turn that would have lateral forces around 1/5th a G to 0.5 G. Lateral acceleration is equal to the centripetal force. V^2/r for 4 fps, this would be 16/r = 0.2 to 0.5 or r of 3 to 1 ft.
I will need to review the clip again. From what I remember, most of the turns were done after it had stopped.
Thanks again for trying this.
31-12-2010 18:52
artdutra04
|
I measured the wheels and the omni wheel seems to bee about 6/7x the size of the outer wheels. Good observation.
|