On board cameras and vision systems and robots! oh my!

[archiver]

Posted by Jon.

Engineer on team #190, Gompei, from Mass Academy of Math and Science and Worcester Polytechnic Institute.

Posted on 5/14/2000 10:58 PM MST



hey y'all,
i'm looking into making a robotcam module that can be moved from bot to bot and i'm wondering how people have handled it in the past.

thanks,
jon

[archiver]

Posted by Thomas A. Frank.

Engineer on team #121, The Islanders/Rhode Warrior, from Middletown (RI) High School and Naval Undersea Warfare Center.

Posted on 5/15/2000 9:38 AM MST


In Reply to: On board cameras and vision systems and robots! oh my! posted by Jon on 5/14/2000 10:58 PM MST:



: hey y'all,
: i'm looking into making a robotcam module that can be moved from bot to bot and i'm wondering how people have handled it in the past.


Dear Jon;

As the team (and I was the person driving the idea) that first fielded a 'RoboCam' at a FIRST competition, here was our development cycle:

1. We used a pair of 'Rabbit' VCR extenders and a low light black and white camera. This was in 96. The Rabbit required 15-18 Vdc, and I was uncomfortable with using main system power (this was the old days of drill NiCd packs, were 2 minutes could very well be all you could expect), so I used six 123 lithium's in series, with a 12 volt tap for camera.
Worked beautifully - the image was a bit bouncy, and there were some dropouts, but it had the 'early NASA' look to it, which was just what we wanted.
The Rabbit's I had worked in the 900 MHz band, so they would no longer be suitable, but at the time were fine.

2. We then upgraded the following year to 1.2 GHz system, and a color camera. This was industrial equipment, we had a weight problem, I couldn't use the on board battery (still used drill batt's), and it wasn't very good, so we bagged it. That was 97.

3. In 98 we upgarded further to a commercial 2.4 GHz system. Single board transmitter and receiver. Used main battery power (first year of the sealed lead acid battery), and a DC-DC converter to isolate the video gear from the noise on the main power buss. Worked like a charm, except in FL, where we had to put the receiver too far away from the field to get a decent signal (it worked OK at the regional).

4. For 99 and 00, we've simply not bothered, as it is not worth the trouble.

Probably not what you wanted to hear, eh? OK, here are my recommendations:

1. Either make sure it has it's on power source, or you use a DC-DC converter to isolate it from the noise on the on-board power buss (which is VERY noise indeed).

2. Get the robot antenna up HIGH, and away from as much metal as you can. Also, make sure it is omnidirectional! You have no way of knowing what orientation the robot will be in at any given time.

3. On the receiver side, use a gain (beam) antenna, and set it up so that its field of view covers the field. Its a cheap way to make the system work a little better.

4. By a copy of the ARRL antenna handbook, and make your own antennas. Much cheaper, and it's not that hard.

5. I've had good luck with equipment from
http://www.mat-co.com
for the transmitter/receiver.

6. Check out Nuts and Volts magazine for adds for cameras. There are literlly hundreds to choose from these days, at very fair prices.

7. While the X-10 system that operated at 2.4 GHz is very nice (I use it at home), it doesn't use an omni antenna, which makes it very hard to use for mobile applications. If you can modify it to have an omni antenna on the vehicle side, it would work pretty well, although the camera is not very low light...

8. and you do need a low light camera for these types of applications. It is surprisingly dark on the field.


I hope this is of some help. Feel free to drop me a line if you have any other questions.

Sincerely;

Tom Frank, Team 121