Sure, the simpler the example the simpler the answer. For example, if you try and animate just a box to look the way you expect Ragdoll to help you achieve, I can hopefully put together an example of what you’re looking for.
It sounds like this is the meat of your question. Generally, it’s real easy to transition from animation → simulation. Because the simulation always starts where you want the animation to finish. From simulation → animation on the other hand is different, because you never know where the simulation might end up. Especially if you expect to tweak any of the simulation settings. And unless you know where the simulation ends up, you won’t know where to start animating from, or what pose the character should have.
There is a semi-related question and answer here.
In it, there are two approaches that apply to this case as well.
- Roll the dice
- Tilt the scale
In your case, you’ll likely best opt for tilting the scale. Namely, to establish a target position you’d like the simulation to reach and gently (or not) pull the simulation towards it. From there, you can use the Blend attribute on the solver to tackle the last 5% of difference between simulation and pose.
Throw together an example of what you’re looking for and we’ll carry on from there.