6000 lumen Deap Sea and Light LED light for our AUV, broad beam, not focused and no reflected tape
I predict we’ll be seeing a lot more of these this year. #teamFlashlight
Why is everyone using such bright lights?? There is absolutely no need for this overkill. Last year we used a really dim (and I mean like a 2 dollar toy keychain light) to illuminate the targets.
If you use a dim colored light the retro reflective properties of the tape will allow you to get a nice HSV range you can use in your programs.
If I’m not mistaken, these teams aren’t using lights for the camera. They’re using the beam of light as the aiming mechanism. If the beam is to the left of the goal, turn your robot till it points to the center.
Thats all we had to test with, we are looking at 1000 lum focused, we are not going to use a computer vision tracking, its for our driver. When he sees the light on the reflective tape he can shot. This will give us an arc in front of the goals
Ah… Mounted Axis Cam with cross hairs on your ds? Why waist time and money putting one of these together?
No camera, just a light. 180 and 25 used these to great success (you know, World Champions and all) last year. No fancy equipment, just point a focused flashlight at the target. They’re really using it more like a laser sight, but big enough that the dot can be seen when it’s on the opposite field wall.
We want our drivers to be looking and paying attention to the field and not the camera. For us its just more intuitive. Never have to bob your head up and down. Just straight.
The drivers won’t see much of a reflection from the retroreflective tape itself. It does a very good job of sending light right back at the source.
During the photon cannon preliminary tests for 4334, I noticed that even with the light pointed considerably high above one’s head, the source of light still makes for a very bright and potentially very distracting element. Especially with extremely high powered light sources such as this one. As a driver, this is something that I am seriously worried about.
Yeah, but the driver will have a camera mounted on the robot, so when the light reflects back on the robot, the camera will catch it and the driver can shoot, especially if the light is a ring light that is mounted on the camera.
Being on the opposite end of 25’s photon cannon last season, I’d have to agree with you.
Seeing more and more of these on the field wont be fun
Nuttyman54 is dead on: 
We tried out a number of Bulb and LED flashlights, that we picked up from Lowes and Home Depot…nothing fancy. We modified them to be powered from 12v-5v and 12v-3v dc to dc converters as needed. We started off with both an automatic vision targeting system (that coupled the web cam and the chassis motors through a PID loop) and the manual targeting of the driver with the flashlight. We had both systems active at the Orlando regional. The driver got so good with the photon cannon ( gotta give George Wallace the credt for that moniker) that he abandoned the auto-targeting. We kept the vision targeting for autonomous - through the practice rounds of the South Florida regional. But, after the driver and operator demonstrated that they could line up the autonomous shots without it, we moved the web cam from up top to down low near the floor. This allowed us to shorten the 'bot to meet the height requirements for balancing from either side of the field, and gave our coach a view of balls in the hard-to-see corners and under the bridges. The 'bot we brought to Championships remained in that configuration.
I find Brandon_L’s comment interesting. During our testing, I found the LED ring, we had for the vision targeting, to be much more distracting, from the driver’s station, than the photon cannon. Do you also find the light rings distracting?
We will be trying both schemes, again, this year, and we’ll see which one proves most effective. Manual simplicity or automated complexity…you learn lessons from both.
Thanks,
Eric
Our driver greatly benefited from have a photon cannon guided system. Never having to look away from the field made it a lot easier to line up. Since our robot only had one shooting range, the light lined up almost guaranteed the shot. It is definitely something to look into if you are having trouble getting the camera tracking to work for you.
Now given the simplicity and the benefits of a light, I wouldn’t be surprised to seeing a lot more teams using a flashlight. I would have to agree that to anyone on the field, this could get annoying fast. Six separate beams of photons will not make it easy on the drivers. Though it could be cool from the stands. I don’t think Dean Kamen ever imagined FIRST evolving into a ROBOT RAVE.
Ooops well then we will have to use the wall above the retro reflective tape…
We used a latest generation Maglight for our tests, and I was seeing spots for about 5 minutes after looking at the beam even from a fairly high angle. I don’t think my eyes are particularly light sensitive, but having a bunch of high powered flashlights on the field could potentially be eye-damaging or at least seriously annoying. I realize that the easy solution is to wear tinted safetys, but I don’t think anyone really wants to do that.
How exactly would you tune the photon gate as your primary aiming system? Would you have the driver look for a certain arc height, intensity, or what? I’ve become mildly interested in using this, especially so the drivers wouldn’t have to hunch over and look on the screen to aim.
To further elaborate on Mac’s point. We were actually wondering whether FIRST would disallow the configuration we were testing as either a safety hazard or an attempt to interfere with the vision of the opposing drivers (as our goals are directly above the opposing drivers). What lights are you guys finding give you the most effective and concentrated beam?
You want to design a targeting system to keep you driver and operator’s eyes on the field! Lining up and shooting quickly is critical to getting a lot of shots on-goal. Leave Dashboards for the coaches, if need be…and for trouble-shooting in-game robot issues.
For an automatic targeting system, a driver/operator needs to be able to engage/dis-engage the system and trust that the software/hardware can line up the shot. Or they won’t use it…period.
In 2012, for a manual system, we tweaked and tweaked the flashlight mount, tweaked and tweaked the shooter angle and speed, and practiced and practiced with the driver to define a “sweet spot” on the target above the basket. Position the robot to hit the sweet spot with the flashlight, and make the basket.
thanks,
Eric
Make sure its not going to blind the drivers, or else you could get penalties…