The impulse-based dynamic simulation is a new method for the simulation of articulated rigid body systems that I have developed during my PhD. The impulse-based method has some advantages compared to classical methods like the Lagrange multiplier method or the simulation with reduced (or generalized) coordinates. First of all it is easy to understand and easy to implement. It can handle all kinds of joints, velocity constraints, collisions and contacts with friction. The iterative method can even simulate models containing closed kinematic chains without additional effort. The simulation provides accurate results just depending on the used tolerance values. Since the computation of impulses is very easy the simulation runs very fast and even complex models can be simulated in real-time. The Lagrange multiplier method has a drift problem and therefore needs an additional stabilization method like e.g. the one from Baumgarte. The impulse-based method targets a valid joint state directly and so it has no drift problem and no additional stabilization is required.

Some older information about this research project can be found here:

Impulse-based dynamic simulation

For more detailed information about the impulse-based method I recommend to read the published papers.