I am tracking a leg during walking, and in doing so only the joints are tracked (ankle, knee, hip). Now, I need to get the center of mass for the two segments (shin and thigh). I have tried to use multiple center of masses (CoM), but two problems arise:
1) I can't subsequently compute a CoM based on the previous CoM's and 2) the masses are located at the extremities, but should rather be located at the center of each segment (this is the standard for this type of modelling).
So I wonder if there's a way I can solve this, or if it could be an option to be able to create a point mass between two tracked point masses - then the mass of the tracked points can be put at a minimal value and the actual mass attached to the new - in-between - point mass? It's a feature I would love to see, which I think would be immensely helpful to anyone tracking multiple linked segments!
Thanks again for your great work, and look forward to a reply. Mikkel
You are right that a CM cannot include another CM--can create circular references and other problems.
The CM track that includes the ankle and knee will behave just like a point mass located there with the total mass of the points. In other words, it does have mass. Is this not what you want? Doug
> I am tracking a leg during walking, and in doing so > only the joints are tracked (ankle, knee, hip). > Now, > I need to get the center of mass for the two segments > (shin and thigh). I have tried to use multiple center > of masses (CoM), but two problems arise: > > 1) I can't > subsequently compute a CoM based on the previous CoM's > and > 2) the masses are located at the extremities, > but should rather be located at the center of each > segment (this is the standard for this type of modelling) > > > So I wonder if there's a way I can solve this, or > if it could be an option to be able to create a point > mass between two tracked point masses - then the mass > of the tracked points can be put at a minimal value > and the actual mass attached to the new - in-between > - point mass? > It's a feature I would love to see, > which I think would be immensely helpful to anyone > tracking multiple linked segments! > > Thanks again for > your great work, and look forward to a reply. > Mikkel
It is close but not quite. If I were to do this, the knee mass would be included in the CoM for both the shin and the thigh - two segments with very different masses. It quickly becomes very problematic unfortunately.
If it was possible to locate a point mass between two tracked point masses - in the same way as a CoM - and then assign mass to this point, it becomes possible to get the actual CoM for a multi-segmented body.
I understand the difficulty but this is the way center of mass is defined in physics. The position of the cm depends on the masses of the point masses.
One thing that might help is to track the knee, then clone that track (Create|clone). This gives you 2 tracks with identical positions. You can give the clone a different mass and make it part of the knee-shin cm while the original is part of the knee-thigh cm. One thing about this is that if you change the original point mass, the clone does not change.
Doug
> Re: Point mass between two tracked points? > > Hi Doug, thanks for the swift reply. > > It is close > but not quite. If I were to do this, the knee mass > would be included in the CoM for both the shin and > the thigh - two segments with very different masses. > It quickly becomes very problematic unfortunately. > > > If it was possible to locate a point mass between > two tracked point masses - in the same way as a CoM > - and then assign mass to this point, it becomes possible > to get the actual CoM for a multi-segmented body. > > > Hope this clarifies it.
