sph-unity

[sph-unity]

We have now created some more scenes to show how the particles behave. In this scene we have a pipe that will spawn particles when the user presses the space button. In this scene we also have a high value for the separation radius which will make the particles stay further away from eachother.

We are pretty satisfied with this version of the project even though we would like to keep doing small changes to improve the behaviour of the particles. 

[sph-unity]

We have now a version of the project that we are quite satisfied with. In this movie you can see how the particles behave when they are dropped from a height. We fixed our algorithm and played around with the different variables to see how the particles would behave. After hours of experimenting, we found a setup that we thought looked good, which you can see in the video. 

A very annoying part of this project has been setting correct values for the variables in the particles. For example we can change the gravityConstant, the restDensity, the viscosityConstant and many other variables to get different behaviours. The paper and other sources don’t really have any recommended values for these variables, so we had to find good values ourselves which took a lot of time.

[sph-unity]

Progress report! We have now done some more changes to the code and our particles actually have some kind of behaviour now. They don’t behave they way we want them to, but they fall to the ground and clump together. We mostly did some bug fixing to make this work.

We ahve also added ridgidbodies to the particles to make them collide with envionments that we create in unity. This is currently working pretty well and it saves us a lot of time because we don’t have to code and bugfix our own collision system.

Now we just have to focus on getting the particles move the way we want them to.

[sph-unity]

First project update!

The paper that we are using for our implementation is “Fluid Simulation using Smoothed Particle Hydrodynamics“ by Burak Erekin.

We have read through the paper a couple of times for our implementation and have now started to translate their pseudo-code and math formulas for the fluid particle simulation.

It has been quite tricky to translate the algorithm from the paper as some parts are not 100% clear and has forced us to look for other sources to read more about details in the implementation.

Currently, nothing works, but we have added code to a particle prefab and we are currently debugging the code.

We are also thinking about how to implement the collision between the particles and boxes/cubes in unity. We might have to follow the paper’s implementation of the collision even if it might be a bit difficult and time consuming to do.