New multi-touch surface planned for this fall

Posted on September 19, 2010 in Hardware

On Monday I will be holding the first meeting of the new Multi-touch Interaction Group (a component of the ACM@UNK club) to begin the process of designing and developing a new multi-touch hardware platform for this fall. I wanted to take some time this weekend to organize my thoughts a little bit and formalize an approach to this project, because time is very important this time around.

Why we’re building a new surface

As some of you may know I have already worked on a multi-touch surface for the CSIS department at UNK, which took two semesters of work to complete. However there were quite a few problems with that surface that prevent it from being very useful or functional in a casual way. The old surface utilized the FTIR technique which in this situation resulted in pretty dim blobs (we lined all four edges of a piece of EndLighten with strips of IR LEDs from Environmental Lights, which I never want to see again in my life). Furthermore we used an off-the-shelf bathroom mirror instead of a front surface, or “hot” mirror, which meant that we had to align the projector and mirror at very unusual angles. Also once we tested out the prototype we noticed it suffered from some pretty severe heat issues which are bad enough to actually trigger the auto-shutdown sequence for the projector! All in all, this prototype was a big headache. Luckily much was learned from this project and the new surface will be a much better prototype than the first was.

Quick overview of new surface and inspiration

The new surface will be built in a radically different way than the original prototype and should  be a large leap forward for us. It will resemble a drafting board and draws inspiration from the mysterious töken multi-touch surface (video on the left). More extensive technical details will follow but the biggest differences between this new surface concept and our current table are as follows: 1) because the entire system in open-air there will be no heat issues, 2) replacing IR LEDs with four IR lasers, which are so bright we can use this sucker outside and 3) because we’re using IR lasers (instead of IR LEDs) the entire system will cost drastically less than the current surface.

Technical details

  • Illumination technique: Laser Light Plane (LLP)
  • Lasers: 4 x 10mW IR LEDs from AixiZ + 120 degree line generating lenses
  • Surface material: Standard glass
  • Webcam: PS3Eye webcam modified by Peau Productions
  • Lasers will be arranged in an overlapping fashion to eliminate occlusion issues.

Also we will likely embed a piezo sensor on the left and right sides of the table so we can gain velocity data