Quote:
Originally Posted by ChrisH
There were also a lot of "noise" issues. IR reflecting of of the field endwalls confused a lot of robots that did get sensors working. I seem to recall our robot tracking a brick wall once. Actually I think we were using an EduBot for a test bed at that point. An auto focusing video camera could interfere as well.
I think the use of coding is intended to help reduce inadvertent interference like that from the video cameras. But that still doesn't deal with reflections and an object does not have to be shinny to be a good IR reflector. The only way to prevent reflections from being an issue is to not give those photons anything to hit.
|
Oh Yeah! I remember 2004, my first year mentoring with FRC. The IR beacons weren't hard to build, the problem was reflected light. The beacon reflected off classroom walls, linolium tile, even the black curtains of the school's stage. The receiver had an auto gain adjustment circuit, but anytime the robot got within 10 feet the circuit was unable to compensate for the IR intensity and the robot was lost.
As far as this little baby's use for this year. First, the operating voltage range is 7-15 VDC. Its definitely an onboard robot device. 7.2 volts = backup battery supply just like CMU and servos, but can also operate at 12 volts just fine so it can be used off main battery source. Also, since FIRST is giving specific idle and in-use power consumption data, this is for your power budget folks! A benchtop power supply isn't going to worry about consumption much. My vote is onboard robot and connected to backup battery circuit, and you'd better watch power consumption especially if this is being used with a CMU and/or servo pan/tilt mechanisms.
The device can receive four different signals, and has an input acceptance angle of +/-30-40 degrees. This is important! Also, the further off axis the sensor is the weaker the signal will be. Will this employ a reliability threshold like the CMU does?
I suspect this could be used for navigation purposes, probably in autonomous. If a device like this were mounted on a pan/tilt a robot could sweep the sensor around and pick up four discrete signals from the playing field, and using servo feedback, could navigate with some reasonable reliability other than the reflection issues. Another possibility, might not be navigation due to the IR reflectivity noted above, but simply to allow the robot to decide on which field orientation it should face. Four input signals = two endzones + two sidelines. Perhaps robot orientation has something to due with scoring or where to score the most points?
Time to mull this over some more....
