Week 15 (May 13th - May 17th)

In this fifteenth week, I began by submitting a Pull Request for the information I had added to the deployment D1 readme, explaining how to launch the smart tutor container.

Additionally, I attempted to launch the RADI container on the Ubuntu virtual machine, but encountered numerous issues. Therefore, instead of testing it on the virtual machine, I launched the RADI from the terminal of my Mac, which worked smoothly.

Regarding the exercises, I reported as issues that the image for the Rescue People exercise wasn't loading and that the Autoparking exercise wasn't functioning.

Throughout the week, I researched how to measure the CPU and GPU values when launching the RADI, discovering a Mac tool called Activity Monitor, from which I could generate different graphs and view the CPU and GPU values. In the end, I concluded that the RADI was launched without graphics acceleration because the GPU was not being utilized.

By the end of the week, I created models for all the exercises in Unibotics, as until now, only the Follow Line exercise had a model. Once all the models were created, I added all the URLs so that depending on the client's location, a different model would be loaded. However, it started to encounter numerous problems when loading the URL from the container. That is, if I launch the tutor locally, there are no issues; everything works as it should. But if I launch the tutor as a container, the models don't load and it responds incorrectly. Unfortunately, I couldn't fix this error this week.

Finally, as usual, I finished the week by updating the blog with another week's worth of content.