This week was spent getting the project far enough for the demo. I wrote the section picking algorithm for runners and biker’s, by implementing a knn influenced machine learning algorithm for picking the section most similar to the runner training images, and picking the section most similar to the biker images (total euclidean distance of closest k points), and going with the closer of the two. I also spent a good bit of time getting my stuff set up on…
DEMO This week we presented our demo, we were able to each demonstrate our individual parts, but we were not far enough along into integration to demo working together. I presented, three lidars working in conjunction, the 3d printed mount, and a very basic lidar detection algorithm Wearable Worked on having the wearable be self-sustained. Meaning integrating the Pi and battery pack into the belt. And wiring the sensors from the chest harness to the pi. (pictured below). Bolted the…
Spent the beginning of the week trying to get bluetooth communication ready to demo, only to realize the framework that I have been using will not be able to work for what we need. I finally figured out the details of the bluetooth peripherals built in to the raspberry pi (its documentation is sketchy at best), and I do not believe there is any way to send a script from the iOS application to start our program on the raspberry…
This week, we heavily focused on our separate parts, as we try to get something cool together to demo. We did get more parts delivered, including the additional LIDAR sensors, so there was a lot of excitement in testing out more of our system and seeing how it measures up to our goals! Here’s a picture of Evan and Will testing out a LIDAR sensor outside on Wednesday.
Spent way too much time working on setting up the app to bluetooth communicate with the raspberry pi, because it turns out (after much digging) that the sample github repo I found will not work for what I need. So I started from scratch, using the CoreBluetooth framework, working heavily with the CBCentralManager class and the CBPeripheral class. Really hoping this will work this time, but not at the point of testing yet. Plan to do more work on Sunday…
This week I came up with my initial approach for features for images (a bunch of blocks of percentage of key points of the images (i.e. 2×4 blocks). The exact number of blocks will be optimized once I have training/test data pictures. After collecting some photos with our camera at about 8m away, I have decided that only a certain section of the image will potentially contain the object we are trying to detect (in the middle of the image)…
LiDAR sensors working together: I was able to get all three lidars to send and receive signals with the RaspberyPi with the TTL converter. I crimped the wires for the lidars so they could easily be attached and detached from the TTL USB converter. Then I edited and debugged the simple python scripts to initialize and wr/rd values to/from all three LiDAR sensors. Using a simple round-robin scheme it prints the strength, and distance values to the terminal of all three…
This week I have really gotten to dive into working on the iOS application, which was nice. Had some bugs to deal with, which took longer than they should have to navigate because this is my first time working with xcode / swift. I’ve also started to look at more demo apps to hopefully speed up this process, rather than trying to learn everything myself from scratch. Also spent a good chunk of time investigating how the bluetooth communication will…
This week we all continued to work individually on our parts of the project. No real changes to our design or schedule. Next week we would like to try to get all of our parts at least far enough for the demo and integrate what we have.
This week I continued to work on and implement the image processing part of the project. I implemented the part to pick the section of the image the object we are trying to detect probably is in, based on which section has the most edges after blurring. I also implemented a k-nearest neighbors classifier. I have also put together the full workflow for the object recognition part (so the only parts that need to be improved are all of the…