Weekly progress 4/18
I made good progress this week. I finally completed a stable function of generating multiple AR objects. I made a lot of changes to the original code and bound the generated items with IDs. And the current code also has good scalability to allow more items to be generated. In addition, I learned MLAPI and completed the online UI. However, it has not been confirmed whether the online function can operate correctly. I am currently using bolts to complete the game mechanics. Since there was no time, I had to make a lot of compromises. But I still try to ensure the rationality of the whole logic to facilitate future optimization. My debugging efficiency is low. This is because most of the functions need to be tested on the phone, and I have to add a lot of debug.log to find the location of the bug. And each packaging takes about 5 minutes, which leads to slower overall development progress. I hope to complete all the basic logic by the end of the project. The most difficult part of the cloud anchor point has been completed. There are also a lot of content that can be optimized, such as easier interaction methods, and better ways to build virtual environments. These parts will be the content of future research. The next possible difficulty may be on the online side. Although MLAPI is similar to the Unreal online framework that I have made, it is still uncertain about the compatibility of Bolts and online functions.
This week I found that the anchor sharing app deployed on the server spent too many resources (exhausted 95% of my free quota). I decided to give up this way of sharing anchor points. I will try to share the anchor point using the LAN.