Wednesday, September 29, 2010

Day Six: Sobriety Test

Poor John Silver the cyborg went out partying and now has been pulled over by the police, who ask Johnny boy to walk a straight line.  Can our cyborg complete this task without waving to and fro?  Yes, yes he can.

The challenge from last class involved modifiying our earlier line following program so that it would waver side to side as much.  We wanted the cyborg to go straight when the line was straight and turn when the line turned.  In doing so, we wanted John Silver to detect the line differently.

We told John to go straight, until one of the sensors touched black.  Then John would turn in that direction, until his sensor hit white.  Then Johnny would go straight again.  Here is what the Pico Blocks looked like.



The simple directions, left, right, straight.

The overall look at the sequence of instructions.


Now here is Mr. Silver walking his straight line, before being released to drive away.  Although when he's walking the sobriety line, isn't he technically driving then?


Friday, September 24, 2010

Day Four: We're Following the Leader...or the Line.

Today's challenge involved teaching our sciborg, John Silver, how to follow the line.  Let's say that he didn't take to this idea.  We struggled for a whole class, trying to get John to notice the line, which involved much beeping/chirping and circular motion.  John did not want to be obediant.

We were confused as to why our program was not working as we desired.  We tried roughly three different ideas, with multiple substitutions within each idea, trying to find the perfect plan, but none of the ideas worked.  John Silver just wanted to play ring-around-the-rosie, by himself.  Here is the version we finally did get to work after much work.



We realized towards the end of class, that John's disobediance may have resulted in part because two wires were not cut down and so touched each other.  Not a good thing.

These two should NOT be touching.



So second day, we fixed our wires and now the program would run.  John would obediently follow a black line marked on a white surface.  All is good.


Wednesday, September 22, 2010

Day Five: Go Towards the Light

"Don't go towards the light!"
"I can't help it, its so beautiful, ahhhh." 

~A Bugs Life

So one of our signed challenges for today and the next class is to get our sciborg, referred to as John Silver, to find the light and approach it.  A simple seeming task at first.  If we just modify the line following to sense light instead....but it turns out to be a tad more complicated than that.

We want the light to be in the center, between the two sensors, which are on either side of our sciborg, almost like eyes.  The light sensors will really rarely want have the same reading, which would equate to the light being in the middle, so instead if the difference between the two sensors is a low number, then that will mean we can go straight.  But what happens if the light is between the sensors?

This is were we sorta' left off.  How to get the sciborg to reposition himself to get the light ahead of him.  After class, I had to do something else, so I practiced my oboe, but then came back to the problem and tried to put the solution of what I wanted in a paragraph.  Sensor one is the right 'eye' and sensor two is the left 'eye.'  What I came up with was If [S2-S1] < 200 go straight;  If [S2-S1] > 200 & S2 < S1 then it will go left; If [S2-S1] > 200 & S1 < S2 then go right.  If the sensors have a difference of more than 200 than the sensors are not equal enough to be going towards the light.  Then if this does happen and sensor one is getting a light reading lower than sensor two it is being pointed directly towards the light and so we need to correct the direction.  Vice versa for if sensor two is getting the more light. 





I'll try this next class and see if we can yet teach Johnny boy to find the light.  He hopefully won't have the same reaction as the moth when nearing the light.

Second Day--- Success. 

Friday, September 17, 2010

Day Three: John Silver, the Cyborg

So our last class, we got to play with these creatures called cyborgs, they were like advanced versions of the previous crickets we worked with.  They are referred to as RDS crickets.  More sensors/output devices could be plugged in.  Anyway, when I heard cyborg I had immediate flashbacks to Treasure Planet and the warning, "Beware the cyborg," so I've decided to refer to the cyborg as John Silver. 

Our first task was to teach Johnny that once he knocks into something, he needs to back-up.  This movement was called The Bounce.  This wasn't too difficult.
The next task involved further developing The Bounce to assist John Silver in escaping a corrall of boxes.  With a few openings, he needed to get out, The Escape.  Here is Johnny in his quest of trying to free himself.



The final task was accomplished in two different ways.  Going The Distance (A) involved counting.  The RDS cricket would count the rotations of the wheels.  So we had Mr. Silver go a certain number of wheel rotations and measured the distance he would travel.  From there we just used a proportion to find how many rotations it would take for John Silver to travel 1.5 meters.

Here is a view of our PicoBlocks.  The number was slightly different as we had just tried





  


to go downhill and stop.


Going The Distance (B) was accomplished, not with counting, but with sensing dark and light.  We had the sensor on John Silver that would sense the darkness or lightness of the color it was facing.  So if we had a dark piece of tape, or light, at the finish line, we could tell John Silver to stop when he sensed the change in light/darkness.  We experimented with the distance between the sensor and discovered the closer to the material the sensor is the greater sensitivity to light.  So this was good and all, but what to do next.



The final sequence we taught John Silver the cyborg was stayin on the table.  If he sensed the edge of the table(the equivalent of darkness) he would stop and retreat from the edge.  This sequence caused some issues for us as we tried to reposition the sensor.  We wanted it to be close enough to sense the lack of light, but we also needed it in the front, otherwise the front wheels would go off and John wouldn't have time to back up before his rear wheels followed his front.  We found the perfect front position for our cyborg and it was a success.



With all the tricks we taught John Silver, its no mystery why he was able to get off the RMS Legacy, with his RDS Cricket programing of course.

Saturday, September 11, 2010

Day Two: A Tale of Two Trains

Friday we modified our night bus train so that it would demonstrate two different concepts of triggering. The wheels no longer function for the night bus. Our turning set of wheels now grace two more muggle motives. Here is a clip of our modified train.




The story that goes along with the movement is that our train driver begins her day's work driving the train/subway in a local city. She goes forward, off to do great things, in the morning. When the sun sets and darkness comes, she goes backward, returning home, where she finds peaceful rest until she rises with the sun the next day and begins her journey all over again.



In this version of our train, the light and motor are both triggered with "edge" triggering. When the light sensor detects darkness, the colors will change into the sunset. When it detects light, the reverse will happen. The motor will start when the button is pushed, reverse when pushed again, and then pushed a third time, the motor will stop. In both of these processes, the action happens when the button is pushed.  Here's a look at what our blocks looked like.

The left column was nicknamed SqW for Squeaky Wheels and is the sequence that turns the motor on and off.
The second column is the Day and Night column.


Now we then also modified our train to show a different kind of triggering, "level" triggering.  Unfortunantly, we did not take a movie of this train, but here is the story (that you can picture in your head) of our level triggered train.



Now we have a train driver that does cross country train driving. She starts in the morning going to her destination, driving until it get darks. When the sun retreats from the sky, she will continue a while longer before stopping to rest. When the sun rises, she continues on her way.



In this cross-country train driver, we have the same color changing system of light detection, but our motor is now triggered by levels. When the button is pushed the train will move, but when the button is released, the train will stop. Unless the button is pushed, our train driver is stuck.  Here's what these blocks looked like and you can notice the difference between this on/off motor sequence and Squeaky Wheels.

The is aptly named the Level sequence, demonstrating level triggering.


So in recap. In "edge" triggering, the button is pushed and the action comences. If the button is released or held down, the action will still happen at the instant the button is pushed. In "level" triggering, the button needs to be pressed down in order for the train to move. The motor is either on or off.



Our train driver would probably like the "edge" triggering system better, otherwise her thumb could get tired holding that button down, the whole day.

Tuesday, September 7, 2010

Day One: The Robot From The Dark Side

Today's creature, nicknamed Chugga, performs like a train wheel.  Yet it will only turn on when the sensor has been sufficiently darkened.  Here is our device in action.
So I guess our wheels would work great on the night bus.

~Kaity