pic: Lucky the Weird Dog

[timtim17]

loading photo ...

[Certified Nerd]

It looks like you weren't able to finish it, am I correct? What are it's strong points?

[Mechvet]

I'm assuming one camera is for driving, and the other is for vision tracking?

Mind detailing on how you've got them both set up?

[GeeTwo]

I see what looks like a Raspberry pi behind the radio, probably for vision processing by the camera in the white housing (has a light ring). I'm surprised the targeting camera isn't connected directly to the pi by USB; the pi seems to have only a network connection. It looks like the RoboRIO was removed for bagging. There seems to be a battery box on the photo right side (near the SRXs), so the RIO presumably fits where you can see a fair bit of belly pan and disconnected wires (near the radio).

I see rhino tread, so presumably a sapper robot as a primary function.

The rollers seem to be designed for a simple intake/return to score low goals. I don't see how they reach outside the frame perimeter, so this may require some precision driving for boulder pickup - especially tricky as I don't see any sign of a camera at that end. I'm guessing the chassis is 26-28" wide to admit a boulder through the gap in the bumpers (here implemented in clear plastic). With the additional width of real bumpers, that may present a problem getting onto the batter.

[D_Price]

The wiring bundles are cringe worthy but hopefully that can get cleaned up

[timtim17]

Originally Posted by GeeTwo

I see what looks like a Raspberry pi behind the radio, probably for vision processing by the camera in the white housing (has a light ring). I'm surprised the targeting camera isn't connected directly to the pi by USB; the pi seems to have only a network connection. It looks like the RoboRIO was removed for bagging. There seems to be a battery box on the photo right side (near the SRXs), so the RIO presumably fits where you can see a fair bit of belly pan and disconnected wires (near the radio).

I see rhino tread, so presumably a sapper robot as a primary function.

The rollers seem to be designed for a simple intake/return to score low goals. I don't see how they reach outside the frame perimeter, so this may require some precision driving for boulder pickup - especially tricky as I don't see any sign of a camera at that end. I'm guessing the chassis is 26-28" wide to admit a boulder through the gap in the bumpers (here implemented in clear plastic). With the additional width of real bumpers, that may present a problem getting onto the batter.

For debugging we had the camera plugged into the pi, but the camera was unplugged when they mounted it. Since our catapult isn't done, it wasn't plugged in again because it's main purpose (auto-targeting towards the high goal) serves no purpose. I don't know much about the bumpers - I'm just a programmer. The roboRIO is under one of the plexiglass shelves. We just drive up to a boulder and it grabs it - there's a camera on the back too.

[whitetail]

Looks like a great robot, Good job!!

[timtim17]

Originally Posted by Mechvet

I'm assuming one camera is for driving, and the other is for vision tracking?

Mind detailing on how you've got them both set up?

We've got a vision camera and two drive cameras - one on front, one on back. The front camera is shown on the driver station by default, but a joystick button activates the back camera, and reverses the drive input so that it's easier to drive relative to the camera view (i.e. to pick up a ball, since it's probably going to be hard to see from the castle).

Our robot and vision code are on GitHub.

[timtim17]

Originally Posted by Certified Nerd

It looks like you weren't able to finish it, am I correct? What are it's strong points?

We're able to go through the low bar, drive over most defenses, take in a ball, and score a low goal. We had some game mechs planned that weren't completed before bag & tag. This is an old picture - this was before our district robot access period.