Navigation on the court

Lucky

I'm currently in the progress of making a tennis ball retriever robot and I was wondering if anybody had some good suggestions as to what would be the best way to track the position of my robot on the tennis court. I've considered GPS, but I have found out that it is not accurate enough for my application. Using line sensors is a possibility, in order to follow the·standard white lines. IR beacons look interesting, but I'm not sure as to how I would use them. Right now, my plan is to use the encoders I have made for my wheels in order to implement dead reckoning. The only problem with dead reckoning though, is that the error in your calculated position builds up over time as your robot moves. Maybe I could use a combination of line sensors with dead reckoning in order to update/ recalibrate the position of my bot. Does anyone else have an opinion as to how to do this better? Does anyone else have some sort of past experience involving a problem like this?

Mike Green

It looks like you've thought this out pretty well. You're absolutely right about dead reckoning needing recalibration fairly often. Line sensors could work, particularly since the environment already has them. IR beacons are also good. You'd mount beacons in the corners of the court, each putting out a slightly different signal so your robot can tell them apart. You'd have a servo motor with the IR sensor(s) so they can be positioned accurately and the sensor(s) would have some mechanism for narrowing the field of view (like a black plastic (IR opaque) tube maybe an inch or two long). You'd periodically scan the environment looking for the edges of the beacons' beams assuming the beacon is midway between the edges of the beam and use triangulation to locate the robot within the space.

icepuck

here's some links that may help get the creative juices flowing..
-dan
robotics.ee.uwa.edu.au/eyebot/
www.seattlerobotics.org/encoder/aug99/ballbot.html

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Arguing with myself--sometimes me, myself, and I don't always agree.
(Former) Caterpillar product support technician

Lucky

Thank you for the reply Mr. Green. How about having two ir beacon arrays strapped to the two poles supporting the net, then the robot having two ir receivers mounted on different servos. The robot could constantly track each beacon and use ASA (angle side angle, with the side being the length of the net) with the law of sines in order to find its position on the court.

I think this is feasible enough. Do you know how far an ir beacon works? (One side of a tennis court is 39ft)

Lucky

icepunk,

Thank you for the reply. I actually had already found one of the sites you mentioned (seattlerobotics-->ballbot), but there is not as much documentation as I hoped about how it goes about tracking its position. It mostly concerns about detecting and searching for the tennis balls, which is one part of the project that I think I have down pat.

erco

Sensing IR outdoors in broad daylight is nearly impossible with hobby-grade electronics. Maybe lasers for short range, if it's safe to use them.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·"If you build it, they will come."

Lucky

erco said...
Sensing IR outdoors in broad daylight is nearly impossible with hobby-grade electronics. Maybe lasers for short range, if it's safe to use them.

Even if you modulate the beacon at 38kHz? My receivers that I currently have contain a built-in 38kHz filter, so it rejects all other frequencies. I thought this was the whole point of modulating an·IR signal so it could reject sunlight emitted IR. Also, wouldn't the tube that Mike mentioned help to block direct sunlight from hitting the receiver?

I like your idea of the laser though. It would probably work from 39ft, wouldn't you agree? I'm guessing reflectors would be wrapped around the posts instead and the laser and phototransistor would be mounted next to each other on the bot. What detects lasers exactly, does a phototransistor work?
How would a laser be anymore indistinguishable from regular light?



Post Edited (luckyshooter353) : 1/19/2010 4:11:18 AM GMT

Phil Pilgrim (PhiPi)

If the infrared sensor gets saturated, as it most certainly will from sunlight, no amount of modulation will get through. It's like, once the needle is pinned, little wiggles just don't register.

-Phil

Lucky

So how about the 2-inch black tube covering the receiver (parallel with ground)? I think that would pretty much stop direct sunlight and sunlight bouncing off the court from reaching the receiver. I think that IR technology has improved, because i'm using IR to detect the balls(Sharp IR distance sensor series), and they work very well and do a good job of filtering the sunlight when I tested them. I'm not sure how they do this though, but i'd be very interested in finding out how.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"You do not really understand something unless you can explain it to your grandmother."


-Lucky[size=-1][/size]

Lucky

I just thought of another possibility! Using magnetic tape and a hall sensor with dead reckoning! Placing a little strip of it every 5 or 10 ft or so to create a grid, then the robot can be programmed to go around to each strip. The grid would be some how programmed into the propeller so that its dead reckoning coordinates could be recalibrated at each strip. The only downside to this though is it requires a relatively high amount of setup in order to work compared to other methods, and you have to take it up each time and buy more tape when your out, etc.

EDIT: I was looking up places to buy magnetic tape and I saw something called magnetic tape recording, which is used in cassettes and videotapes like you put in VCRs. Does anybody know if this could be taken out and used successfully with a hall sensor?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
"You do not really understand something unless you can explain it to your grandmother."


-Lucky[size=-1][/size]

Post Edited (Lucky) : 1/19/2010 6:25:52 AM GMT