Using this degree project as our final project in SI2335, Simulation Physics, turned out to be a great success and it can not be stressed enough how much time and pure workload we saved in doing this. We finished the course with an outstanding presentation and also made a lot of progress in this project.
During the days leading up to the final presentation, which was held on tuesday this week, a lot of effort was put into refining the algorithm we already had and creating scenes in which we could simulate different scenarios. In addition to that, we worked quite a bit on the report that was, of course, a part of this project and we realised, upon its completion, that we now had our first draft of the report associated with this degree project.
The simulations we ran are shown in the videos below and the results are quite interesting. Something that hits you right away when looking at these videos is that the movement of our crowds show striking similarity to fluid flow, which would be expected considering the method we are using. You may also notice that things seem to be running pretty smoothly now. It is, unfortunately, not because the algorithm has been optimized but because we render the videos in a different way now. Instead of recording the runs with a screen recorder we save each and every frame as png-files, by simple scripting, and use an external software to turn these individual frames into videos. The benefit is that we can produce videos at a fixed frame rate, of our choice, regardless of the frame rate at runtime. The result is quite pleasing.
After the presentation, we decided to take a break for a couple of days and so we have not done anything since. Although, the project is resumed with the submission of this post.