Back to school after the holidays, we ran tests to validate our idea for detecting other Phyllos with IR radiations. Quick reminder, we had ordered two powerful IR transmitters (LTE-R38381S-ZF-U and SFH 4441) and several receivers, some in AGC2 (TSOP2256 and TSOP4856) and others in AGC4 (TSOP4456). Signal management is a little different depending on the type of gain control (AGC): some receivers are more suitable for noise reduction and other lower detection times.
We will briefly go through the IR detection protocol once again.
First of all, Phyllos give themselves a unique identifier by communicating via Wifi. Once each Phyllo has an identifier and knows all existing identifiers, it starts the detection protocol.
In the order of their identifiers (ascending order for example), the Phyllos take turns turning on their IR transmitters. The neighboring Phyllos detect it using their rotating receiver and associate it with an identifier. After that, the Phyllo turns off its IR transmitter and it is the next Phyllo’s turn.
With Xavier, I tested the feasibility of this protocol on Monday. Here is our experiment:
We turn on the IR emitter with a square wave signal at 56kHz:
As a first test we toggled on or off the emission of this signal every 500ms, to see if we could manage to correctly detect when it is and isn’t being emitted. We didn’t really have any trouble getting a coherent signal from the receiver, so far, everything was great.
The next step was to narrow the IR receptor viewing angle. As previously discussed in this article, we will detect the direction of another Phyllo by finding the angle at which the rotating receiver starts to detect an IR signal, and the angle at which it ceases to receive that signal. The direction of the Phyllo will then be computed as halfway between these two angles.
Using this method, the viewing angle of the receptor shouldn’t matter, as long as it’s symmetrical, but we figure having a narrower view can only improve the precision. Our IR receivers have a viewing angle of approximately 90°. Tarik helped us put thermo-retractable sheaths on the receivers in order to reduce their angle of detection to a few degrees.
By quickly moving the receptor so that the emitter source passes through its field of vision, we managed to detect the two edges of the signal received : the moment when the receptor starts to see the signal and the moment when it stops seeing it. Those edges then triggered Interrupt Requests, which we can use to compute all the angles we need.
Finally, we chose to use the most powerful transmitter : LTE-R38381S-ZF-U. We plan on emitting from inside the Phyllo, and using the sculpture to diffuse the signal in all directions. We confirmed this would work by emitting through one of our 3D-printed test petals : we could still detect the signal. We plan on doubling the power by providing two transmitters per Phyllos, as it is difficult to predict the impact of the diffusion surface, that is to say the impact of the shell with the petals in its entirety.
For the receivers : AGC2 receivers are recommended for fast transmissions, AGC4 are recommended for noise reduction. We therefore think we will chose the two receivers and will probably use AGC4 for the diffusion of the common time reference (as explained in this article) and AGC2 for the detection of Phyllos.