Ruofei Du, CS PhD

Max Potasznik, CS Masters
Kent Wills, CS Masters

Supplementary Material (Kinect Side):


Zen gardens are a peaceful way to visualize space and tranquillity. Our project seeks to connect the zen garden to the room in which it sits; imbuing the garden with properties of the room's occupants. In this way, the zen garden becomes not just a visualization of an ideal space and tranquillity but of the actual tranquillity of it's surroundings.

2014-04-16 01.22.21_副本.pngScreenshot 2014-04-16 01.22.44_副本.png
Screenshot 2014-04-16 01.23.11_副本.pngScreenshot 2014-04-16 01.37.15_副本.png


  1. Finalize design for two motors working together prior to gear/track production.
    • IMG_1930.JPG
  2. Begin production of gears and tracks. This took a lot of trial and error regarding 3D printing before we could reliably print gears and tracks. Custom Pinion for Motor.
    • 86rqx.gifIMG_1946.JPG
  3. Assemble track and test gearing.
    • track.jpg
  4. Design and protoype motor-carriages that can move along the geared track.
    • 86s68.gif
  5. Assemble frame for full XY table.
    • full.gif
  6. Calibrate motor movement with tracks to determine range of motion and create API for setting sphere location.
  7. Kinect
    1. Use Kinect to detect XY position of people in a room, regard the nearest player to the Kinect as the primary user.
    2. Calibrate Kinect field of view and feed Kinect coorinates to the XY table.
    3. Configure the response time of recorded sensor readings in relation to actual motor movement, currently the magnet's position in the zen garden is updated every 500 milliseconds.
      • sandflow.PNGsandflow2.PNG
  8. Determine Paint is not the best medium to use for drawing
    • Paint.jpg
  9. Settle with Sand
    • sand.jpg

More Video and Photos


Iteration Highlights

IMG_1958_副本.jpg2014-04-08 19.09.32_副本.jpg
2014-04-10 18.39.09-1_副本.jpg2014-04-10 18.38.55_副本.jpg2014-04-07 18.24.11_副本.jpg

2014-04-09 18.04.39_副本.jpg2014-04-09 18.03.54_副本.jpg

Challenges and Solutions

We went through many different 3d printed designs. It was a challenge to create a rack and pinion (gear) that worked seamlessly together. We ended up scratching our original rack and pinion design for a design we found on thingiverse, this however, did not solve all of our problems. Once we printed the gear, we realized that we needed to fit the pinion precisely on the gear. At first, we thought that this could be accomplished through drilling out a slightly smaller hole in the gear to fit the motor. This ended up not being ideal, because the fill being around 10% made it so that the drilled hole could not keep the pinion centered on the motor shaft, thus producing a wobble. In the end we combined Anis's (student in the bunny group) form fitted design with our gear in TinkerCAD and we simplified the overall design of the gear to produce a quicker and more reliable print. With countless failed prints, ~4 pinions @ 2 hrs each, and ~5 racks @ 3 hours each, we were looking at more than more than 24 hours spent at the 3D printer. Lessons learned include: simple designs and 10% infill are reliable and can decrease print time and spend the time to ensure your 3D model is correct to reduce the amount of time for re-prints.

K'nex were crucial to our design process. One of the great things about children's toys is the fact that they can help you iterate through many different prototypes quickly to see what works. If we would have been working with more permanent materials like wood and glue, we may never have finished our project.

The overall design that we came up with for an XY table is slightly lacking. The gears have potential to slip (need higher torque motors) due to the strength of the magnets on the design (4lbs of force per magnet), the design probably has an upper limit for speed because the motor must transport itself along with whatever it is carrying, unlike a belt system that you would see on the maker bot, and the frame can only withstand a small amount of weight on the top, so we had to resort to using very light materials.

Reflections and Future Work

Due to the time spent on the construction of the X/Y table, we did not have much time to explore the different types of motions or sounds that we wanted to capture. In the future we wish to explore these functions thoroughly:
  • Visualize the noise level in the room: The magnet marble would render a circle with a radius proportional to the sound amplitude in the room.
  • Determine if abstract translations of the atmosphere can produce "beautiful" or "meaningful" drawings.
  • Re-explore the use of drawing with paint instead of sand.