 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
DOES ANYBODY OUT THERE HAVE IK LEGS?
I mean, does your hip levitate off the ground by some sort of magic, and
your legs occasionally pick up your feet to move them along and drop
them? Kind of like a flying car that keeps dropping sandbags, lifting
them, and throwing them ahead as it glides along.
OR ARE YOU LEGS FK?
That is, is your hip structurally supported off the ground, while one
foot basically stays put on the ground due to friction, while muscles
rotate your thigh and shin bones about their joints, and the position of
your hip is at the mercy of a complicated function of all these
variables? Is the position of your knee backsolved by the postion of
foot and hip or is it solved by the rotation of your hip alone?
My legs, at least, are FK! ;-)
This animation is a walk cycle of my blobmon involving a true FK based
walk cycle. No foot slippage, by definition! Am I right in that this is
a significant accomplishment? With this kind of cycle, I could have a 4
or 15 jointed leg, which could be difficult to solve with IK math. The
position in Y is determined by locating the "bottom" of the figure with
2500 trace calls from -y, and a trace call from y to the floor, so I
could have it walk up a gently sloping mountain.
It would not have been possible without various features of Mega 0.4. I
was greatly disappointed, however, that the init_spline funciton does
not work at all (It only works in cases where you could have just as
easily used the sine function). I spent days tweaking my cycle before I
figured this out. The motion may not be as smooth as possible, because
I'm stuck with a linear spline until I can figure this out. I may be
forced to write my own cosine-spline macro to get what I need!
Hopefully, this figure doesn't look like a big booger! ;-)
Post a reply to this message
Attachments:
Download 'thebodywalker11.mpg' (409 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <38ef51f8@news.povray.org>, "Greg M. Johnson"
<"gregj;-)56590\""@aol.c;-)om> wrote:
> It would not have been possible without various features of Mega 0.4. I
> was greatly disappointed, however, that the init_spline funciton does
> not work at all (It only works in cases where you could have just as
> easily used the sine function). I spent days tweaking my cycle before I
> figured this out. The motion may not be as smooth as possible, because
> I'm stuck with a linear spline until I can figure this out. I may be
> forced to write my own cosine-spline macro to get what I need!
Have you tried using the other spline patch available in MegaPOV? The
one described in section 5.9 of the MegaPOV docs?
--
Christopher James Huff - Personal e-mail: chr### [at] yahoo com
TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
Web page: http://chrishuff.dhs.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
As I showed in p.b.i, I guess the #init_spline is actually working, as it
gives a very smooth curve which visits all of the points you give it. Thus,
my earlier complaint is INCORRECT. It was surprising, I must say, that I was
getting values > +60 when my points ranged from -120 to -10.
Chris Huff wrote:
> In article <38ef51f8@news.povray.org>, "Greg M. Johnson"
> <"gregj;-)56590\""@aol.c;-)om> wrote:
>
> > It would not have been possible without various features of Mega 0.4. I
> > was greatly disappointed, however, that the init_spline funciton does
> > not work at all (It only works in cases where you could have just as
> > easily used the sine function). I spent days tweaking my cycle before I
> > figured this out. The motion may not be as smooth as possible, because
> > I'm stuck with a linear spline until I can figure this out. I may be
> > forced to write my own cosine-spline macro to get what I need!
>
> Have you tried using the other spline patch available in MegaPOV? The
> one described in section 5.9 of the MegaPOV docs?
>
> --
> Christopher James Huff - Personal e-mail: chr### [at] yahoocom
> TAG(Technical Assistance Group) e-mail: chr### [at] tagpovrayorg
> Web page: http://chrishuff.dhs.org/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Greg M. Johnson" wrote:
> This animation is a walk cycle of my blobmon involving a true FK based
> walk cycle. No foot slippage, by definition! Am I right in that this is
> a significant accomplishment?
It impress me, but I have a little trouble with the moving speed:
walk usually tend to keep the same amount of momentum for the whole body.
In the case of a human body, the biggest mass is the upper-body
(from the waist and above). In your animation, this part accelerate and
decelerate too much (your walking blobmon will be more easily tired than
a regular moving thingy :-)
The balancing of arms is also used to keep the vertical position of the
whole mass center (barycentre ?) moving along a line, as much regularly as
possible: when the waist go down because of the increased angle of the legs,
the arms raise (well, they rotate around the shoulder, but in a vertical
projection, they raise!), thus displacing upward a little the center of
the mass relative to the body, trying to keep it on the straight line of
the move.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Very impressive. I can see a lot of hard work went into this one - how
did you do it?
"Greg M. Johnson" wrote:
>
> DOES ANYBODY OUT THERE HAVE IK LEGS?
> I mean, does your hip levitate off the ground by some sort of magic, and
> your legs occasionally pick up your feet to move them along and drop
> them? Kind of like a flying car that keeps dropping sandbags, lifting
> them, and throwing them ahead as it glides along.
>
> OR ARE YOU LEGS FK?
> That is, is your hip structurally supported off the ground, while one
> foot basically stays put on the ground due to friction, while muscles
> rotate your thigh and shin bones about their joints, and the position of
> your hip is at the mercy of a complicated function of all these
> variables? Is the position of your knee backsolved by the postion of
> foot and hip or is it solved by the rotation of your hip alone?
>
> My legs, at least, are FK! ;-)
>
> This animation is a walk cycle of my blobmon involving a true FK based
> walk cycle. No foot slippage, by definition! Am I right in that this is
> a significant accomplishment? With this kind of cycle, I could have a 4
> or 15 jointed leg, which could be difficult to solve with IK math. The
> position in Y is determined by locating the "bottom" of the figure with
> 2500 trace calls from -y, and a trace call from y to the floor, so I
> could have it walk up a gently sloping mountain.
>
> It would not have been possible without various features of Mega 0.4. I
> was greatly disappointed, however, that the init_spline funciton does
> not work at all (It only works in cases where you could have just as
> easily used the sine function). I spent days tweaking my cycle before I
> figured this out. The motion may not be as smooth as possible, because
> I'm stuck with a linear spline until I can figure this out. I may be
> forced to write my own cosine-spline macro to get what I need!
>
> Hopefully, this figure doesn't look like a big booger! ;-)
>
> ------------------------------------------------------------------------
> Name: thebodywalker11.mpg
> thebodywalker11.mpg Type: MPLAYER2 File (video/mpeg)
> Encoding: base64
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thanks
0. body: one big blob with arms & legs as rotating
cylinders & spheres.
1. hip/knee/foot rotations: linear splines set up to enable keyframing
2. foot staying in one place: vtransform of foot position
3. vertical position: 2500 trace calls from -y to positions on grid
between min_ & max_extent x & z
Yes, it's not the most beautiful thing, but it actually works like a real leg.
We use "IK" because it's a neat math trick, but the body really works as FK,
does'nt it?
Lewis wrote:
> Very impressive. I can see a lot of hard work went into this one - how
> did you do it?
> > OR ARE YOU LEGS FK?
> > That is, is your hip structurally supported off the ground, while one
> > foot basically stays put on the ground due to friction, while muscles
> > rotate your thigh and shin bones about their joints, and the position of
> > your hip is at the mercy of a complicated function of all these
> > variables? Is the position of your knee backsolved by the postion of
> > foot and hip or is it solved by the rotation of your hip alone?
> >
> > My legs, at least, are FK! ;-)
> >
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I meant, how did you do the FK?
"Greg M. Johnson" wrote:
>
> Thanks
>
> 0. body: one big blob with arms & legs as rotating
> cylinders & spheres.
> 1. hip/knee/foot rotations: linear splines set up to enable keyframing
> 2. foot staying in one place: vtransform of foot position
> 3. vertical position: 2500 trace calls from -y to positions on grid
> between min_ & max_extent x & z
>
> Yes, it's not the most beautiful thing, but it actually works like a real leg.
> We use "IK" because it's a neat math trick, but the body really works as FK,
> does'nt it?
>
> Lewis wrote:
>
> > Very impressive. I can see a lot of hard work went into this one - how
> > did you do it?
> > > OR ARE YOU LEGS FK?
> > > That is, is your hip structurally supported off the ground, while one
> > > foot basically stays put on the ground due to friction, while muscles
> > > rotate your thigh and shin bones about their joints, and the position of
> > > your hip is at the mercy of a complicated function of all these
> > > variables? Is the position of your knee backsolved by the postion of
> > > foot and hip or is it solved by the rotation of your hip alone?
> > >
> > > My legs, at least, are FK! ;-)
> > >
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
The hip is the origin.
The thigh rotates about the hip, with one point on the origin.
The shin first rotates about its knee rotation, with one point on the origin. Then
it is translated down a distance equal to the thigh, and rotated again.
Lewis wrote:
> I meant, how did you do the FK?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
