Toy Train Sequencer

Team Members
Alexis Taylor, Kelly Park

Design Progress

We have addressed some concerns that arose in feedback to our midterm presentation. One of these concerns is about working with actual sound. This is the next step that we plan to take, and first we must decide where to place the speakers. One option could be to mount them on the train, but that will add bulk to the [moving] train part. Another option is to mount the speaker in the station, which may end up being the more viable option.

Another concern was about complexity. So far, our hopes for users to be able to reassign sound classes by interacting with the station may be too complex, especially since our main target user base is children. We can assign fixed sounds to the main tokens, which will limit the full range of functionality for the set, but is ultimately easier to implement for now. We would like to retain the recording token, however. We would also like to prototype the train = melody idea, and the idea that the speed of the train controls the speed of the melody (and moving the train backwards will reverse it) by adding a simple tiny music box apparatus to the train.
The movement of the wheels can turn the crank, which in turn will produce a melody which can speed up or slow down (and play backwards) with the movement of the train.

The communication between the tokens and the train seems like our hardest challenge. We thought about RFID, but it will not work in the way that we want (ie, temporal sequencing of tokens) because we envision tokens being recognized by the train as it passes them, regardless of near/far difference in where the token is placed. However, the RFID signal transmitted from the train radiates outward, not in a straight, precise beam. Token passive signals will not be read in the order that maps naturally to the train moving around the track and passing each token (or lines of tokens).

One possible solution is to use photo sensors: mount an LED to the train and photo sensors to the tokens. However, this still has the problem of light radiation not being in a straight line (unless we want to use a laser and potentially blind our users), and the possibility that some tokens will block others from receiving the light unless we constrain the token placements within the fixed lane.

Another solution would be to create a fixed surface for the train and tokens. The tokens will sit in slots wired in groups, like lines radiating from a center point outwards. All of the lines will be wired back to the station, which will house the speaker and microcontroller. When one particular point of the train crosses each line, it acts as a switch and closes the circuit, activating all of the tokens placed in the slots on that line at the same time. It solves the problem of temporality, but reduces the ability for users to place tokens where ever they desire. (It also closely mirrors the way that many top of the line train sets already work - the ones that have moving or light-up parts that are activated when the train passes.)

Lastly, we would love to get more advice on our implementation.

List of Materials

1. Toy Train Set - a train car, a set of track, tokens(objects)
2. Train Table(plate)
3. additional plate to create slots or token lane
4. Arduino Uno
5. Breadbox
6. Music Box (http://www.amazon.com/Kikkerland-Hey-Jude-Crank-Music/dp/B00421KJZO/)
7. Conductive wires (for wire mapping)
8. LED for train
9. Photosensors for tokens
10. speaker
11. wires