Kinect usage at competitions

[jyh947]

Originally Posted by BigJ

We joked around about having our human player go in and pretend to shoot baskets while the robot ran autonomous, but decided we didn't want to accidentally misrepresent ourselves to the judges

This thought actually crossed my mind after Kettering.

[Grim Tuesday]

We use the kinect on the robot for it's infrared camera.

[Chris is me]

There is so much strategic potential with the Kinect that isn't really being utilized. A lot of teams are either using it as the only method of controlling their robots or don't use it, which is a little silly.

HOT's paper explains that they use the Kinect as a safety override. That is a great use.

Say you miss shot 1 - use the Kinect to adjust for shot 2. Maybe you are trying to go to the middle bridge, and someone beat you to it - use the Kinect to cancel and drive to where you want to start teleop instead of breaking your manipulator or whatever.

[DominickC]

Team 23 was the only team to use the Kinect at the WPI regional.

[StAxis]

Only one team legitimately used it once, however 79 Krunch realized about halfway through Orlando that it was a great opportunity to have someone dance through the first 15 seconds for the rest of their games

[JakeD]

1540 and 2443 both used it in Oregon. 1540 was first pick by the 1 seed and used it to time when they would drive toward the bridge. If they got beat to it they would just stop and not risk driving under the bridge and ensuring it rolled any leftover balls off the bridge. I'm not positive what 2443 used it for but they were still using it all the way through the finals.

[Schnabel]

Originally Posted by DjScribbles

...I am a bit suspicious that this could be due to cable length...
..but it had an atom...

I think the atom probably is a bigger issue [/atom hate].
I know that the cable FIRST is using is designed specifically for the Kinect with a powered repeater, so that would explain why the hub didn't work for you. Sorry you had so many issues though!

[DampRobot]

It seems to me like the Kinect was never a very good idea. Not only does it not allow for precise control, but it is probably worse than a good autonomous program. Was there a similar lack of enthusiasm for 2008's remote control thingy that could be used during autonomous?

Our team isn't currently using the Kinect. We considered doing a "shoot now" type of thing, so we didn't overlap with other robots shooting. I like HOT's idea, but part of me believes that trying to correct a first missed shot using the Kinect would only result in a second air ball.

[BigJ]

Originally Posted by DampRobot

It seems to me like the Kinect was never a very good idea. Not only does it not allow for precise control, but it is probably worse than a good autonomous program. Was there a similar lack of enthusiasm for 2008's remote control thingy that could be used during autonomous?

Our team isn't currently using the Kinect. We considered doing a "shoot now" type of thing, so we didn't overlap with other robots shooting. I like HOT's idea, but part of me believes that trying to correct a first missed shot using the Kinect would only result in a second air ball.

The difference with 2008 was that the randomizer determined where balls would be for a match after robot placement. There is a "similar" (but not very) aspect here with the timings of your alliance partner's shots, but an adjustable autonomous routine and communication pre-match can resolve most of those problems.

[Jared Russell]

The 2008 IR remote was very important for one reason: The Randomizer. Game pieces were put in random locations prior to each match, and the IR remote was a way to tell your robot where they were.

Rebound Rumble has no equivalent aspect of randomness. Sure, you might get beat to the bridge, or the balls might be placed slightly different on the bridge, but imagine if there were 2 more "wildcard" balls that were placed somewhere on the floor pre-match. THAT would have been a strong incentive to use the Kinect.

[DavidGitz]

Here is a copy of my post on this from another thread (which references another thread discussing the same thing...)

Originally Posted by DavidGitz

You can see my other post here but in short we found the use of the Kinect very valuable. We did not have the time or resources to implement a closed-loop driving method, so we chose to use the Kinect to drive the robot during the Autonomous period along with other modes that used simple open-loop control.

The main benefit that we found at the KC Regional was that with the Kinect the Kinect Driver was able to react much more dynamically to certain situations, such as lining up for the coopertition bridge to knock the balls down, picking up a random ball that didn't make it into the hoops and primarily to steer around other robots that were executing their autonomous driving modes. Initially we had programmed it to allow the Kinect Driver to be able to shoot balls that were missed during the initial autonomous shooting but with such a limited time that was ineffective. I wasn't able to watch every one of our matches but as far as I know we were able to get to the Coopertition Bridge and tip it to our side before our opposing alliance in every match.

[DjScribbles]

Originally Posted by Schnabel

I think the atom probably is a bigger issue [/atom hate].
I know that the cable FIRST is using is designed specifically for the Kinect with a powered repeater, so that would explain why the hub didn't work for you. Sorry you had so many issues though!

Well, when we actually attempted to use the kinect, we used a dual core laptop, which kept up fine.

I'm still quite suspicious that it was a cabling problem. I don't recall reading hearing about a repeater; I thought Microsoft simply tested with the cable length first used and gave them the nod for using it.
We were able to drive it around for about 5 minutes in our build area, and we had what seemed to be much better control.

Has anybody else using the kinect observed similar issues to mine?

[Dkt01]

Originally Posted by Schnabel

I think the atom probably is a bigger issue [/atom hate].

I would agree with your suspicions. In practice the classmate yielded delayed results for our team, but using a more powerful laptop greatly increased the response time. It will be interesting to see if this holds true on a real competition field.

4096 plans to use the Kinect at the Midwest Regional next week.

[RufflesRidge]

Dj,

Personally, I would be much more suspicious of the physical environment than of the cabling. The Kinect is fairly free to move around inside that black box that it is housed in, especially as robots collide with the field. It is also pretty easy to ratchet the Kinect servo so it is tilted up or down instead of straight ahead.

If you plan to continue trying to use the Kinect at your next event, have your Kinect operator make sure the Kinect is centered, pulled out towards the edge of the box (so the box does not block any of the image) and level.

If your laptop can handle it without trouble, you may want to run the Kinect Server in debug mode (the shortcut is in the FRC Kinect Server folder) to verify what the Kinect is seeing at the start of each match. If you need the screen real estate you can hide the debug mode window behind your dashboard after verifying that the Kinect is positioned properly.

[kws4000]

We are definetly planning on using it this weekend at GVSU.

->5-speed shooter control-right arm (0,.2785,.35,.59,1) and conveyor operation-left arm (off-off, on-off, on-on).
->Last but not least, Pretty Light control w/ Head buttons!

We tested it yesterday and it works like it should! But we can also operate w/ old boken timing code from Gull Lake by flipping z-throttle on joystick before match begins.