23-02-2005 21:00

David Guzman

Re: pic: 820's finished product

It sure does look like a tripple play robot,
what kind of drive train does it have?

Very nive machine
I love the neons

Dave

23-02-2005 21:09

Arefin Bari

Re: pic: 820's finished product

definitely gotta agree with you on "Very nice machine"

they posted a picture earlier... they are using omnidirectonal drive. its pretty cool...

good luck team 820...

23-02-2005 21:19

Bleric

Re: pic: 820's finished product

Do you mean a holonomic drive with 3 omni-wheels? I've always wanted to see one of those! If that's what it is, how are you controlling it?

Again, that is one great looking robot.

23-02-2005 21:59

dlopez

Re: pic: 820's finished product

I really like the unorthodox shape and cant wait to see the omni drive

23-02-2005 22:02

JKnox

Re: pic: 820's finished product

Thanks for the praise. It's so beautiful I'm starting to miss it already. This year we decided to control our robot with joysticks There is a lot of programming involved, but basically, our robot can move forwards, backwards, sideways and spin like crazy on the spot. Our arm can also stack on both goals quite well.

23-02-2005 22:24

408Oxidation

Re: pic: 820's finished product

Loving the black tetra, very nice...
Odd shape, but it looks like it would work. VERY unconventional
I love it

23-02-2005 22:27

JKnox

Re: pic: 820's finished product

We can build robots, but tetra's are way out of our league as you can see.

23-02-2005 22:31

Arefin Bari

Re: pic: 820's finished product

Thats what I am guessing...

Anybody from 820..???

Please fill us in. thanks.

23-02-2005 23:11

JKnox

Re: pic: 820's finished product

We are using a holomonic drive with 3 omniwheels. If you're wondering how it works then I can tell you...
The wheels are equidistant from the center of the chassis and are each respectively perpendicular to three lines at 120 degrees. Simply put, each omniwheel is parallel to one of the small side plates on the chassis.
If we want to move in a straight line, since the omniwheels have very little lateral friction, movement is simple vector addition.

+ / = | ( and / being 2 omniwheels and | being the direction)

For instance, if we want to move forwards or backwards in a straight line, the two omniwheels located under the support beams of the arm move at the same rate. If we want to move sideways the vector addition becomes quite a bit trickier as you can imagine. The programming is entirely based on the calculated rates that each of the motors should spin at, but I'm not a programmer so I don't know those details.

As for actual controls for the robot...
We have one joystick for sideways and forward/back movement and when we hold the top button we spin based on the direction and force that we apply to the joystick (left or right).

24-02-2005 01:02

shyra1353

Re: pic: 820's finished product

Wow! That is insane and amazing and spectacular and <insert every other synonym here>. I can't even begin to imagine the calculations the programmers must have had to go through.

Question though. Why did you choose that sort of drive train besides the obvious reasons of control and maneuverability?

Good luck this season! (A part of me is glad we won't be facing you in competition and another part of me wishes we were so I could see her in action)

24-02-2005 01:06

thoughtful

Re: pic: 820's finished product

WOW!, that is one sweet robot. I did see your earlier posted videos and the robot spins like crazy. It will be a great advantage for you guys to maneuver in any direction you want.

Only question, Did you guys try pushing against other robots? how is the pushing power of this baby?

24-02-2005 08:25

tiffany34990

Re: pic: 820's finished product

WOW!! way cool drive train!! how fast are you guys at??? speed and spinning away fast surely seems like a good. but this is a great drive chain system-- i hope the students can explain well and get a design award. good luck!!!

and GREAT JOB AGAIN!!!!!!!!

24-02-2005 08:33

JKnox

Re: pic: 820's finished product

Maneuverability and control were the only strategic reasons for us to build this type of roboot. Although, we also built this type of robot because very few teams have tried this type of chassis before. It was also somewhat of a challenge which makes the build period that much more fun. The robot is entirely student built except for the omniwheels (courtesy of AndyMark) and one tiny sprocket mount for the motor for the arm. I would also like to mention that we don't have a shop at our school and only had one day to work at an actual machine shop on a weekend.

I'm pretty sure the video you saw was not of our robot, but our robot CAN move like that.

As for pushing power... our robot is very weak when it comes to pushing power because the wheels aren't perpendicular to the motion. However, we built this bot based on the idea that if you can avoid pushing matches you won't need pushing power. It may be weak, but we also have sideways pushing power which most other robots don't have

24-02-2005 09:26

Goober!!!

Re: pic: 820's finished product

neon lights look great!!! and cool robot!!!

24-02-2005 14:13

JKnox

Re: pic: 820's finished product

I'm pretty sure our robot goes 7 ft/s, any direction, but don't quote me on it.

24-02-2005 18:22

alphastryk

Re: pic: 820's finished product

why did you decide on three wheel holonomic instead of 4?
we did 4 this year for ease of construction/programming. it seems to be adding complexity and losing ease of construction to use 3, but you probably had a good reason...what?

24-02-2005 19:22

JKnox

Re: pic: 820's finished product

In retrospect, it probably would have been easier and more effective to build a 4 wheel holomonic drive. The main reasons why we chose the 3 wheel design were: looks, weight and cost. The 3 wheeler also fits the triple play theme and our logo (the deltatech triangle) so it was fate. These may not be very "good" reasons, but like I said... it was fate.

Personally, I proposed the 3 wheel design because I was tired of building box shaped chassis's and I love to watch programmers go nuts.