Jeff’s Status Report for 03/30/2024

What did I do this week?

This week I worked on testing the A* algorithm, further developing and improving the web application, and worked with my team to test the localization system.

For testing the A* algorithm, I wrote a few scripts to help me construct a simple graph utilizing the png image of a floorplan. From here, I was able to do some performance testing to ensure the algorithm would be able to function on these larger (though simple) graphs. The algorithm was able to execute in around ~10 ms on my machine, which seemed to be a fairly positive result.

I worked on improving the web application. The first thing I worked on was to convert the server from using asynchronous HTML requests to instead using websockets. The main motivation for this was to improve the server’s performance to decrease the number of requests (offsetting more of the load onto a regis server) and decrease the latency between updating the user’s position on the browser. To implement the conversion, I worked on utilizing Django Channels and implemented the code necessary to facilitate communication between the tag and the user. The improvements over using HTML requests were fairly significant. Firstly, websockets seem to result in significantly faster update frequencies–from my testing, it seemed to be an order of 10 times faster. Additionally, the latency from sending the update to being updated on the user’s browser was improved dramatically, at around 5 times faster. With these changes, I also rewrote the script for the tag to send messages to the browser.

I worked on further development of the web application, moving a significant amount of hard-coded values onto the server’s database, as well as improving the user interface. For the floor view, I added a small form floating on top of the page for the user to input their desired destination. Then, I added a “connecting” indicator that shows up whenever the browser is no longer connected to the tag device.

Finally, I helped test the localization system with my team by mapping out the A-level corridor in HH and setting up a floor on the webapplication to view the user’s location.

Is my progress on schedule?

My main goal for this week was to assist my team in improving the accuracy and stability of the localization system. My improvements in the web application’s update frequency directly helped with the responsiveness of our tag’s localization, and I was able to implement the user interface necessary for displaying the user’s position, so I believe I was able to achieve this goal. Altogether, my progress is on track.

Next week’s deliverables:

The midpoint demo is approaching and I believe the localization system is sufficient in providing a stable calculation of the user’s position. My next deliverable is to start working on the localization system. My main goal is to have the webserver be able to take inputs from the user of where they would like to go, run that information through A* algorithm, and generate the paths to display on the browser.

 

Leave a Reply

Your email address will not be published. Required fields are marked *