DrawVinci - A Robot that Produces Unique Works of Art
Archbishop Carney Regional Secondary
Floor Location : M 014 D
For my Grade 8 Science fair project, I made a robot that draws unique works of art called The DrawVinci. This type of a machine is known as a polargraph and works using stepper motors, gravity and basic math. Changing the pen type, grid size and speed as well as uncontrollable attributes, such as pen shake, creates images that are different each time. DrawVinci works by dragging the pen from one coordinate to the next. Moving the pen to the coordinate involves calculating the number of turns the motors must rotate to let out the precise length of cord from both of the sprockets. This process relies on Pythagorean theorem and requires two calculations for the two triangles representing the pen's coordinate. I chose this project because I have an interest in mechanical engineering, electronics, mathematics, and programming.
I set out to solve the problem of 'how can simple motors and gravity be used to create a realistic drawing machine.' This was done through the building of the machine itself. When I conducted my main experiment I applied my hypothesis 'If I use the smallest grid size, the slowest speed, and the finest pen tip size, then the drawing produced will be the most realistic.' To analyze the "realism" of the images, I designed a questionnaire for students/teachers to complete. A sample of 53 people were surveyed in total. One overall question was asked, of the eight images which did survey participants think was the most realistic? I then had to think of a way to isolate the importance of each individual variable so I added a series of 12 questions, pairing images against one another to isolate variables. This let me see which variables were most important for determining realism.
I encountered a few problems/challenges during my procedures. While I thought speed would have made a major difference in image, it did not. The largest difference seen on the high speed images was a slight, almost unnoticeable looping on the tips of the infill, however, it did affect the voting results. My hypothesis was proven to be true after I saw that when the 'low speed' (print speed was set slower), 'fine pen' (a smaller/finer pen tip size was used), and 'small grid' (the pixels were at their smallest size) settings were enabled, the most realistic image resulted. The grid size and pen tip type did impact image votes however. The finer pen size won over the thicker one even if the image had a large grid size or was 'less realistic' than their thick penned counterparts. The favorite image out of all eight was number 4 which was determined to be the most realistic image (low speed, small grid, thin pen). This vote was of course subjective as it was asking for an opinion of realism. Yet still after all these votes, the best and most realistic one turned out to be the image whose variables closely matched my hypothesis.