I've been working on a 3D rigid body engine for the past while as a pet project, and so far I'm happy with how it runs. I've got a world that responds correctly to collisions between convex rigid bodies, using GJK w/EPA and the sequential impulses method.
The next step I would like to take is to implement some fracture simulations into the engine, but I've hit a bit of a brick wall. Research and reading first came up with http://graphics.berkeley.edu/papers/Parker-RTD-2009-08/Parker-RTD-2009-08.pdf which is a paper on how the DMM (Digital Molecular Matter) engine works, which is far more advanced than I'm aiming for currently.
I went down the rabbit hole, and ended up back in 1999 at this paper: http://graphics.berkeley.edu/papers/Obrien-GMA-1999-08/(which it seems is the basis for the DMM engine anyway, with some modifications).
I'm trying to implement this method, however am struggling with it. Has anyone here any experience with this topic, or does anyone know of an implementation of this (or a similar paper) that I could use to get started?