2012 course Things That Think (CSCI 7000) at The University of Colorado - Boulder.
The story of our bug-catching spider automaton goes like this: A giant evil spider resides
in a big web. One day an unfortunate lady bug gets trapped in the web and the scary
spider pounces on it.
We came up with the idea for this six-week project together and discussed implementation
possibilities at length, particularly with respect to moving the spider and detecting the bug.
For example, although we ended up using IR sensing for bug detection, we considered
several other possibilities including touch sensors and image processing. At that point, the
highly modular and reusable nature of the project components allowed us to build
separately, and we are posting our Instructable as a two-part series.
Part 1, described in this Instructable, shows how to build the following elements:
* The frame to which everything is attached
* The spider web and associated lighting
* The spider mover, which is an XY table beneath the web
Part 2 shows how to build the remaining elements:
* The bug detector
* The bug
* The spider
Readers who would like to build this project should be aware of the following limitations:
The XY table moves nicely under human power, but not with the servos, and we think this
outcome is a result of the materials not being sufficiently rigid. It's possible that adding
another servo to move the lower slider from both sides would be helpful, too. Here are
some alternative Instructables XY tables to try:
Low Cost Hobby Servo XY Table
Internet Arduino-Controlled T-Slot XY Table
DIY CNC Router
Our intent was to be able to display the web in any position, such as leaning against a wall.
For this reason, we used strong magnets on both the spider and the mover. Although this
idea worked well with models during pretesting, the final spider clings too tightly to the
web and doesn't move well. For this reason, we would recommend a different spider /
mover magnetic connection, as well as stronger servos to help overcome drag.
The spider's orientation is fixed. We had discussed using a rotating arm on the spider
mover, with the pivot point toward the front of the spider, but did not build it due to time
limitations. This arm would have allowed the spider to turn as it moved forward along an
arc traced out by the spider mover.