Dr. Sprague thought that one possible option would be to join Nhung on her project!!! I’m really excited to be working on it as well since there really is not much or not any research on this specific area. Since we both are targeting the same problem, but through different mediums, I thought it would be a good idea to join her and work together on this project.
I started off by reading some articles about bats and how they perceive the space around them through echo location. I’ve looked at presentations that explained how echo location works in bats, but there was no previous research on how to reproduce that in a device.
The purpose of the project is to create a device that can be trained to uses echo location to perceive depth (using neural networks). As of now Nhung built up the neural networks and successfully found an architecture that gives a pretty good representation of depth through echolocation, but of course not completely accurate as of now.
I have also started to focus more on coding, and especially finishing exercising with numpy and starting on Keras. For now my main focus is to create a program that can chop off spectrograms so that they all start at the same time. For that I am using python and numpy. As of now, I am using data from a previous experiment to create the program, but later it will be helpful for the actual project. Afterwards, I will practice coding neural networks.
Since we are in need of more and better data collection for our bat ears, team Sprague is travelling to Virginia Tech to learn more about it this Friday!!!