| Topic: |
Science > Physics |
| User: |
"" |
| Date: |
22 Jul 2006 12:26:25 AM |
| Object: |
Rigid body simulation |
I am trying to make a small rigid body physics simulator, and I have
come to a roadblock. To get the right motion of a rigid body is easy
if you store it's angular and linear motion quantities (displacement,
velocity, and accelleration) seperately. However, I want to only store
the linear components of each particle seperately.
Idealy, I would calculate the seperate quantities of motion from the
combined ones, then integrate the motion, and finally combine them
again for the next iteration. However, I get confused because most
articles, and even textbooks, aren't specific when they talk about
torque, angular momentum, etc., as to wheather they refer to those
quantities for the whole system, or the individual particle. Any help
would be much appreciated.
.
|
|
| User: "John C. Polasek" |
|
| Title: Re: Rigid body simulation |
22 Jul 2006 09:08:10 AM |
|
|
On 21 Jul 2006 22:26:25 -0700, wrote:
I am trying to make a small rigid body physics simulator, and I have
come to a roadblock. To get the right motion of a rigid body is easy
if you store it's angular and linear motion quantities (displacement,
velocity, and accelleration) seperately.
You'd be surprised how difficult it is to get the "right motion of a
rigid body". A rigid body has an inertia tensor with 3 principal axes
and if it's not spinning on one of those axes there will be a crazy
wobble you never imagined.
However, I want to only store
the linear components of each particle seperately.
If you wantto derive the rotation from the linear components, take
half the curl of the velocity FUNCTION. It has to be a function,
mathematically.
Idealy, I would calculate the seperate quantities of motion from the
combined ones, then integrate the motion, and finally combine them
again for the next iteration. However, I get confused because most
articles, and even textbooks, aren't specific when they talk about
torque, angular momentum, etc., as to wheather they refer to those
quantities for the whole system, or the individual particle. Any help
would be much appreciated.
John Polasek
.
|
|
|
|
| User: "Ben Newsam" |
|
| Title: Re: Rigid body simulation |
22 Jul 2006 03:45:12 AM |
|
|
On 21 Jul 2006 22:26:25 -0700, wrote:
Idealy, I would calculate the seperate quantities of motion from the
combined ones, then integrate the motion, and finally combine them
again for the next iteration. However, I get confused because most
articles, and even textbooks, aren't specific when they talk about
torque, angular momentum, etc., as to wheather they refer to those
quantities for the whole system, or the individual particle. Any help
would be much appreciated.
If you design your classes carefully, then it should become apparent
which properties apply to the object and which to the system. For
instance, assuming you are modelling a system in which time is an
overall constantly elapsing quantity, each object will have among
other properties, a position (x,y, and z), but time will be a property
of the whole system. As you analyse what your classes should be, then
the properties will reveal themselves. Hint: do the analysis before
you start coding.
.
|
|
|
|
| User: "" |
|
| Title: Re: Rigid body simulation |
22 Jul 2006 09:33:46 AM |
|
|
See the Airplane Simulator writeup at
www.berkeleyscience.com
All the details are there, including an implementation.
tavianator@gmail.com wrote:
I am trying to make a small rigid body physics simulator, and I have
come to a roadblock. To get the right motion of a rigid body is easy
if you store it's angular and linear motion quantities (displacement,
velocity, and accelleration) seperately. However, I want to only store
the linear components of each particle seperately.
Idealy, I would calculate the seperate quantities of motion from the
combined ones, then integrate the motion, and finally combine them
again for the next iteration. However, I get confused because most
articles, and even textbooks, aren't specific when they talk about
torque, angular momentum, etc., as to wheather they refer to those
quantities for the whole system, or the individual particle. Any help
would be much appreciated.
.
|
|
|
|
|
Related Articles |
|
|
