Introduction: Urban Wilderness Monitor

Goal:
The goal of AndyBot 9001, or the Urban Wilderness Monitor (UWM), is to collect quantitative data about various species living in an environment through animals interacting with a series of sensors.

How:
The UWM possesses a number of sensor inputs that, when triggered, add to counters and release the door of a feeder. The counters represent different information, while the food is meant to serves as feedback to encourage further participation from the organisms. In this way, a feedback loop is created between humans and their subjects. An UWM benefits the human user by providing information about the preferences and abundance of local wildlife. It benefits subject by offering a tasty reward.

Designers:
Larry Smith and Mary Macheski-Preston. We made this project for our Cybiotic Interaction Design Class taught by Andrew Quitmeyer at Georgia Tech. For more information, visit the course blog at http://lmc.gatech.edu/~aquitmeyer3/cybiotic/ !

Step 1: Supplies

Materials Used:

x1 Force Sensitive Resistor - Square
x1 Mini Photo resistor
x2 HS-322HD Servo
x1 Arduino
x1 Bottle
x1 10,000Ω resistor
x1 100Ω resistor
Squirrel/Bird friendly food
T-shirt Fabric
100 ft of wire
Craft foam
Header pins
9 volt battery
Alligator clips
Electrician's tape
Scotch tape

Step 2: Planning

Creating the Design:

To start off, we planned out the basic interaction loop for our project and concluded that we wanted two different sensors (one for detecting birds and one for detecting squirrels) and a servo-operated feeder. We tested many kinds of sensors, including a PIR motion sensor, a proximity sensor, a flex sensor, a stroke sensor, a photo resistor, and a pressure pad.

During this process, we used a stuffed squirrel with the correct weight and light hand gestures to imitate our target animals. In the end, we concluded that a pressure pad with a weight threshold high enough to eliminate birds would be the best for detecting squirrels and a well-placed photoresistor would be best for detecting the presence of birds. We also tested out several different bottles and various methods of hanging them to help define our feeder design.


Final Inputs and Outputs:

Inputs:
Squirrel-> Pressure Pad
Bird-> Photo resistor
Output:
Servo-> Dispenses Food

Step 3: Construction

Setup:

The UWM is designed to be integrated into a tree and therefore does not have much more to its physical form. Its body is composed of three parts - the feeder, the brain, and the arms.

First, we created the arms, or "core" of the UWM by creating a cable out of several wires - one for 5 volt power, one for ground, and three to control the photoresistor, pressure pad, and servo. Then we soldered each of the formerly mentioned components to their own independent 10ft cord and attached it to the cable. This allowed us to place the sensors or servo independently in the tree. Lastly, we improved the pressure pad's performance by attaching craft foam to it and embedding it in fabric so that it could be easily tied to a branch.

At the opposite end of the cable, we attached each wire to a header pin that plugged into the appropriate Arduino pin. The "brain" of the UWM consists of a protective box covered in waterproof tape with an Arduino inside. The Arduino is either powered by a 9 volt battery or plugged in to a laptop (a computer is optional and can collect extra data, but is not necessary).

The UWM’s feeder is composed of an upside down bottle, the servo's arm, and a servo. The servo is either set up close to the bottle or attached to it, and the arm, which we made out of cardboard reinforced with plastic and attached with hot glue, blocks the bottle’s mouth. When a sensor is triggered, the arm moves and allows food to be dispensed via gravity. It is hung using a metal bracket wrapped around the bottle and with metal wire. 

Step 4: Code

Attachments

Step 5: Final Product

Step 6: Future Plans

Future Plans:
Future plans for the Urban Wilderness Monitor involve creating a more waterproof body and experimenting to find a better way to attract animals to the sensors than food. The reasoning behind this is that different squirrel and bird populations are exposed to different kinds of food. For example, the squirrels on Georgia Tech campus were entirely indifferent to bird seed, whereas the squirrels in north Atlanta suburbs will break into garages for it. One could either trainer the animal to accept the desired food before installing the UWM, or use different methods of attracting subjects such as bird calls.

Other Ideas:
The UWM could be modified to monitor competition more than two different species or even within a single species in different environments.