More line tracking videos

Here are videos showing the line tracking using the new sensors from the kop. We used the lines with the actual line width and length. We demonstrated choosing the left line, the right line and straight.

http://falconroboticsteam842.org/2011linetraking.aspx

Looks good! We hope to be at that stage tomorrow.

Now, I am expected to wire the photo sensors myself… Do I just solder them and use heat wrappers? Or is there a less “permanent” way to wire it up?

You can probably just twist wire connections together and cap the connection.

This is 1646’s linetracking kitbot. We built the mostly standard kitbot and attached two of the photoswitches to the front.

We wired power and common (brown and blue) to a 20 amp breaker on the PD board and spliced signal from LO (white) to a PWM cable which got plugged into DIO ports. From there it was just some code to keep a short memory of what sensors was last on the line so we could try to turn in the correct direction when we lost it.

If you have any other questions about how we built it I’d be happy to answer them but it is nothing special.

we attached our three sensors with 3 bolts and a cardboard box. We calibrated the sensors and loaded the default LV line code. We were able to follow a straight line for 4 seconds (2 feet) and then it keeps turning off. Anyone been playing with the Labview code?

Are the LEDs toggling as expected when passing over the line? I found it useful to leave the robot disabled, and simply push it to pivot over the lines. Turn the trim knobs on the sensors if necessary using the LEDs to identify the value of the sensor.

The tutorial at http://decibel.ni.com/content/community/first/frc?view=all#/?tagSet=1001 in the lower left will lead you through some things to think about.

The stop after four seconds is happening because the code interprets three sensors true differently before four seconds and after. After four seconds, it assumes the robot has reached the T and it halts. Before four seconds, it assumes it is the Y or a sharp angle and keeps going.

Also, if your robot has only driven two feet in four seconds, it sounds like the gearing is far different from the kitbot. You may need to adjust the forward speed, the Y value to make better progress.

Greg McKaskle

Has anyone tried using the official carpet? Per the rules the official carpet is S&S Mills, Sequoia 20 Oz. Level Loop Pile, Color: Ground Pepper. Has anyone even found this carpet? I was not able to find it on S&S’s website. From the color photo in the game manual the carpet appears to be a light or medium grey color.

My concern here is that all these videos show contrasting color line following (i.e. light tape, dark carpet), but according today’s team update the gaff tape is also grey. The color photo in the game manual makes the greys appear to have some contrast.

In any case teams may want to consider making their line following easy to modify. Or try testing line following with grey on slightly darker grey.

How far apart are you placing your three sensors?

CENTER and +3 inches on each side or so?

thanks!! You guys look like you really have this working great!!

Call Debbie at S and S Mills…1-800-241 4013. We ordered the FIRST standard ground pepper (grey) carpet, 20 oz, Olefin commercial loop carpet and it arrives tomorrow for our new field. We bought two 60 ft.x 15 ft. rolls for $1500 total (includes shipping). Be sure to identify yourself as a FIRST team.

We already have carpet for testing and can’t afford a full new field. Can you take some pics when the carpet arrives and post them here?

Thanks!