{ design process}

Although we had a long and lengthy list of about twenty different ideas we could've pursued for our final project, we came out with two winners!

1. Elephant Art (which we did): Make a robot that can, on voice command, draw specific images using paint and paper. Dress the robot as an elephant with the ar of the robot being the trunk. Program it to make noise, dip a paintbrush in paint, and paint three to five pictures.

2. Spider Spelling (which we were thinking of doing): Like in Charlotte's Web", make a robot that cna spell out letter in thread when verbally told letters. Build a long straight path with pegs liing the sides so the robot can hook thread around the begs to spell our letters--and afterwards, wind the thread back up as well so it can begin again.

So, we decided to tackle Elephant Art. We began with planning Plan A, Plan B, and Plan C, which ended up evolving into a hybrid of Plan ABC with no backups. We started with building the base, which was originally meant to hold the whole elephant's body but turned into being just big enough for the elephant's head. (Our robot was bigger than we anticipated.) We experimented with gearing until we came up with a 3:1 gearing that worked. This was so that Todd could draw horizontally. Then we began building motion modules for the different paint strokes we were planning to use. Going online to the PicoCricket Website, we used motion module "Chomper" for the mouth, and the "Spinner" for the trunk spinning. We came up with our own motion module to raise the "Spinner" up and down.

{ the chomper: mouth goes up and down }

{ the wheels give Todd left and right motion so he can draw horizontally; the circular motion module lets his draw circles!}

{ the tracks allow Todd to draw up and down strokes}

 

Tweaking all these motion modules and coming up with spatial placement on the limited size base we had made took us a long time. We just could not figure out how to raise the "Spinner", and after we did, we need to know how to attach the three modules together. After finally putting things together, we added sensors and began testing. It turned out that paint and paintbrushes were just not going to work, because we hadn't programmed Todd to dip the paintbrush in paint periodically. So we used mini-markers instead. We also had trouble coming up with what simple drawings we should pre-program into Todd, and decided upon the smiley-face, "YUM", and himself.

{attempt with paint!}

{mini-marker yum!}

Thing began to work (if only sometimes) and it was then that decorating and dressing up Todd and his habitat became a necessity. The night before exhibition, as Sue furiously programmed, Christine made the background and made Todd look like an elephant--er, elephant head. The IR sensoring wasn't working so after staying up until 4 am programming, we came back early the next day to work on things some more. Thankfully, things really came together about half and hour before exhibition time.

______________________________

There were many problems that we had to deal with and change, one of the biggest being our entire gear train and tire combination the night before exhibition day, and the creation of a moving easel that would communicate with Todd via and IR sensor and receiver. Although programming wasn't too complicated, it had to be very precise to render any sort of recognizable picture. And the robot was extremely temperamental, working one moment and not the next--we had a lot of trouble with the gearing for the motion modules (if they somehow broke free they would go berserk), and ended up having to hotglue a lot of Lego pieces together. Most of what we had planned had to be changed to something else; for instance, we were going to have the audience interact with Todd so that he would draw what they told him to do, each different action by the audience being accompanied by a specific brush stroke. Along with this we were going to have Todd know how to draw three drawings so he could make art for the audience or the audience could make art with him. We also considered having Todd sense a lot more, such as when his brush is going to go off the page, or what someone is telling him to draw, but as time ran out we couldn't implement these things. Sometimes the rubber bands on the motors moving Todd left and right would fall off, sometimes the gears would get stuck, and while we were building, axles would bend, the sensors would fall, parts would break off, etc. However, exhibition day it worked pretty well--our "YUM" was the biggest success. :)

______________________________

 

Click here for our code!