I started off this week with a rather concerning realization, which is that our original plan to use four microphones was flawed. We intended to connect two microphones to each of the Jetson Nano’s two I2S ports, but this would not allow for us to distinguish the microphones’ signals from each other and analyze them individually, which we need to do in order to determine the direction of the source. I talked about it with the team, and we have two options moving forward:

1. Stick with I2S and just use one microphone on each input, for a total of two audio inputs.
2. Use the remaining three I2C ports on the Jetson (Heather’s using one of the four already for the motors), which would mean getting new analog microphones and some analog-to-digital converters.

I have been proceeding with the former option, since I believe two microphones should still be adequately accurate for our purposes. After all, that’s how our ears operate! In addition, as a backup, I spent a good chunk of time researching analog microphones and ADCs and ordered some. They should be arriving by the time I’m done implementing the two-microphone array (one of my goals for the coming week), so I’ll be able to test that setup and verify whether it’s accurate enough before deciding to make the change to the I2C three-microphone array. We’d prefer not to have to switch to the latter option for the same reason why we chose the Adafruit I2S MEMS microphones to begin with, which is that analog isn’t as nice to work with, and these I2S microphones have digital output.

In the meantime, I’ve done the math/physics to calculate the angle of the source from the center of the two-microphone array given the time difference between when the sound reaches each microphone and the distance between the microphones. I also wrote out the pseudocode for it. I would be writing and uploading actual code to my Jetson, but it only arrived a few hours ago (it is currently the evening of 10/10).

Lastly, I’ve been working on the design review slides and preparing, since I’ll be the one presenting this time.

My goals for this week are to get the Nano up and running with the two-microphone setup so that I can begin its acoustic localization capability and then determine whether we need a third microphone. I will also work with the team on the design review paper.