This week I worked on parsing location messages from the wearable device. The wearable device encodes messages slightly different than messages sent by another pi, because messages from the wearable devices are all sent using raw bytes rather than a string converted to bytes. I …
This week was very exciting for me. In addition to working out some major kinks in my wifi connectivity, I was finally able to connect to Jeffrey’s test server that was set up on the RPi4. From there, it was only a little while longer …
I have been having carpal tunnel issues for a while, and this week it aggravated beyond control so I had to visit the doctor. I was given a hand brace, to help posture my wrist better. This has affected my efficiency as, I can’t type well with the brace on, as it prevents my wrist from bending.
That being said, we spent the Monday of this week working on the Design Review Report. And after that got back to our capstone project again. I had been looking into ways to track the movement of the same object over different pictures. The best way, I have found so far to do this is to use: https://www.opencv-srf.com/2011/09/object-detection-tracking-using-contours.html, on OpenCV. I wrote a base implementation for this, but I need to add it to the RPi and check if it works with the pictures as I want it to or not.
Also, this method does have the possibility of providing false positives. And hence needs to be tested over different objects, but so do the other methods like using meanStdDev to check if the standard deviation of two frames is similar, and if they are associates them to be the same object.
This week I set up a remote desktop on the Raspberry Pi Zero W with a VNC viewer, to be able to see the input from the camera. And started taking pictures of the platform we have designated as our initial clutter test zone. Initially, …
This past week was a great time of development on the embedded systems side of things! We hit a couple milestones. I finally got communication working smoothly with the breakout board. The Wi-Fi driver is tentatively complete. I was able to successfully use the driver …
During the start of the week, I worked on the math behind the accuracy of our Wi-Fi positioning method so that we can put it in our design report. I was actually sick for the first half of the week so I didn’t get as much work done as I would have liked, since I also had to make up work from a lot of other classes. But I continued the making the hub server multithreaded so I can handle the communication between multiple clients at once.
This week we finished doing some research to write our design report. Our camera is now able to receive input to the Raspberry Pi Zero. The hub can now handle multiple clients at a time and recognize which type of device the client is. We …
This week I worked on building the server to connect with multiple clients and setup a protocol between the hub server and the different clients (button, wearable, camera). To remember what devices are correspond to what, I created a file that will store mac addresses …
This week I was working on the design review presentation and working on the algorithm to outline the object detection. For the design review presentation, I made specific slides for each component that we have and updated it with the progress my teammates had made so far. And then prepared a script and prepped for the presentation on Monday.
Besides that, on the computer vision, I made progress on image processing and am able to detect object contours, from various methods I found out online.
So there were lots of events that took place over the past week. First of all, I got to spend many hours looking over an oscilloscope and debugging SPI communication with the Nucleo board. Since the Wifi board does not have documentation on the format …