Sounds like a marvelous trip! I hope you both enjoy it!

I'll look forward to your write up -- gives me time to chew on the concept
meanwhile! (I think I did have a not-quite-correct understanding of it
initially, and am starting to see it more clearly now as you explain how control
would work.)

- Robin


--- In HexapodRobotIK@yahoogroups.com, "David Buckley" <david@...> wrote:
>
> Robin
> I am going to be away from tomorrow for some time, myself and a fellow
researcher, Reuben Hoggett are taking a trip round Europe visiting places and
people important in the history of Robotics from the forgotten period before
about the mid sixties, plus some art robots from a little later. Reuben has been
the main researcher for my History pages and I can't keep up with him!
> I will write a discussion outline of Be+Have, hopefully something clearer(?)
and post it to Randy's group when I get back.
> David
>
>   ----- Original Message -----
>   From: Robin
>   To: HexapodRobotIK@yahoogroups.com
>   Sent: Monday, May 18, 2009 4:05 PM
>   Subject: [HexapodRobotIK] Re: Robotic Leg for Medical Research
>
>
>
>
>
>   --- In HexapodRobotIK@yahoogroups.com, "David Buckley" <david@> wrote:
>   >
>   > Robin
>   > Thank you for the paper link.
>   > Back then it wasn't as easy to search for papers and in the intervening
time I haven't bothered as this is the way I want to do it. I never imagined for
one moment that my idea was unique
>
>   It sounds like it could be very unique, though! I just meant that I may be
mis-understanding!
>
>   Did you want to move the discussion to Randy's list? I'll wait in that case
until you've had a chance to start it there!
>
>   - Robin
>

(cross posted to HexapodrobotIK list)

A Lynxmotion quadruped?  There had been some attempts in the past, but none
really as successful as the LM hexapod.  You might start seeing some soon.
The major missing part has of course been the software.

> -----Original Message-----
> On Behalf Of Peter Balch
>
> <snip>
> legs are better than crocodile legs disappears. The torques in the joints
> are much the same. It's not "obvious" that a mouse is better than a lizard
> or that Little Dog is better than a Lynxmotion Quadrupod.

Oh, I'd say Little Dog gets around a lot better then any current LM
Quadruped!

> Sprawled quadruped robots have a wider stance and so are more stable and
> many can walk equally well in any direction. Little Dog style quadruped
> robots may have a longer reach and may fit through narrower gaps.
>
> Which is better overall? Darned if I know. Maybe Little Dog's designers
> have
> spotted something I haven't.
>
> Peter

I'm beginning to wonder about that.  I need to draw the triangles of support
and all that, but in some experiments with my quadruped chassis (not TRUE
four legs); I found that it quickly fell over when I kept one leg up.  I
rotated the front two legs to face forward from their 34 degree radial, and
ended up with a long stance.  But by having the front two legs facing
forward (and angling them in tandem to do translation) I got my balance
basically back.  The two rear "pseudo legs" (wheels) were in the same
sagittal plane as the now forward-facing front legs.  Very little actual
time "walking" with this set-up, so the facts may not all be in.

This leg arrangement might give up good translation for stability.  Perhaps
LD did that as well?  I'd love to get the dimensions on the leg components,
joint freedoms, and their arrangement on LD!

Alan  KM6VV

> -----Original Message-----
> From: theinnerbreed@... [mailto:theinnerbreed@...]
> Sent: Wednesday, May 27, 2009 3:57 PM
> To: arachner_robotics@yahoogroups.com
> Subject: [arachner_robotics] Quadruped Gaits.
>
> All quadrupeds use one or more of the following gaits
>
>       Walk
>       Amble
>       Trot
>       Gallop (rotary & transverse)
>       Canter
> ------------------------
>
>
> Gait Description
> A simple description of the timing of a
> particular gait requires the following
> information
>
>       Number of legs
>       Gait period
>       Duty factor & step trigger for each leg
>
>
>
> Gait Phase
> We can think of the gait phase a value that
> ranges from 0 to 1 as the gait cycle proceeds
> We can choose 0 as being any arbitrary point
> within the cycle (such as when the back left foot
> begins its step)
> The phase is like a clock that keeps going round
> and round (0.1, 0.1, 0.1)
> For a particular gait, the stepping of the legs and
> all other motion of the character can be
> described relative to the gait phase
>
> Step Cycle
> In one gait cycle, each individual leg goes through a
> complete step cycle
> Each leg's step cycle is phase shifted relative to the
> main gait cycle
> The step cycle is broken into two main stages
> Support stage (foot on ground)
> Transfer stage (foot in the air)
> The amount of time a leg spends in the support
> stage is the support duration (& likewise for
> transfer duration)
> SupportDuration + TransferDuration = GaitPeriod
>
> Duty Factor
> The relative amount of time a foot spends on the ground
> is called the duty factor
> For a human walking, the duty factor will be greater than
> 0.5, indicating that there is an overlap time when both
> feet are on the ground
> For a run, the duty factor is less than 0.5, indicating that
> there is a time when both feet are in the air and the body
> is undergoing ballistic motion
>
> Step Phase
> The step phase is a value that ranges from 0 to
> 1 during an individual leg's step cycle
> We can choose 0 to indicate the moment when
> the foot begins to lift (i.e., the beginning of the
> transfer phase)
> The foot contacts the ground and comes to rest
> when the phase equals 1 minus the duty factor
>
> Step Trigger
> Each leg's step cycle is phase shifted relative to
> the main gait cycle
> This phase shift is called the step trigger
> The trigger is the phase within the main gait
> cycle where a particular leg begins its step cycle
>
> ---------------------------------------------
> Quadruped Gaits
> Quadruped: 4 legs
> Walk
> The basic slow gait of most
> quadrupeds is the walk
> Very slow walks may involve
> 3-4 legs on the ground, but
> normal walks involve 3 legs
> on the ground with a brief
> moment with only 2
> The duty factor is therefore
> relatively high (.6 ~ .Cool
>
> Amble
> Ambles are like a quicker version
> of the walk, but are also
> associated with larger, slow
> moving quadrupeds
> The duty factor is often in the .5 ~
> .7 range, but some quads amble
> at even lower duty factors
> Elephants use the amble gait
> exclusively. The front and back
> legs are often very close in phase
> (shifted by around .1 or so)
> The gait often involves a
> noticeable swinging of the body
> from left to right
>
> Trot
> The trot is a medium paced gait
> where alternate diagonal legs
> step nearly in sync (though often
> slightly led by the forefoot)
> The duty factor is usually
> relatively low (<.4) and there are
> moments where all 4 legs are off
> the ground (actually, cats
> sometimes trot at a higher duty
> factor.)
> Before Muybridge, most horse
> trainers believed a trotting horse
> always had at least one foot on
> the ground
>
> Canter
> Unlike the first 3 quadruped gaits
> we looked at, the canter is
> asymmetrical
> The canter is a medium speed
> gait, but a bit irregular and not
> usually used for long intervals
> Some horses canter as they slow
> down from a gallop
> Sometimes, the timing of the
> canter is more like .6, .0, .0, .1,
> with 3 legs stepping in rapid
> succession, alternating with the
> 4th leg
>
> Gallop
> The gallop is the fastest
> quadruped gait
> The gallop involves an alternation
> between the front and back pairs
> of legs, but slightly out of sync
> There are several subtle
> variations on gallops, but they
> are generally separated into
> transverse and rotary gallops
> Horses tend to prefer the
> transverse gallop, as do most
> other quadrupeds

--- In HexapodRobotIK@yahoogroups.com, "Alan Marconett" <KM6VV@...> wrote:

> I'm beginning to wonder about that.  I need to draw the triangles of support
> and all that, but in some experiments with my quadruped chassis (not TRUE
> four legs); I found that it quickly fell over when I kept one leg up.

Hi Alan,

Yes, in general, the better the geometry is for efficiency and agility while
walking, the less stable it will be when stationary. You may want to experiment
to find the boundaries for static stability then go just far enough those to
have a comfortable margin of safety.

I can only guess here, but I presume that's why BigDog so often "prances" rather
than stands still. It *can* stand still, because it rises from a camel crouch,
but since it lacks feet (except in the beta version :), it's probably at risk of
slipping when standing, even though all fours are on the ground.

- Robin

Hi Robin,

> -----Original Message-----
> On Behalf of Robin


> > I'm beginning to wonder about that.  I need to draw the triangles of
> support
> > and all that, but in some experiments with my quadruped chassis (not
> TRUE
> > four legs); I found that it quickly fell over when I kept one leg up.
>
> Hi Alan,
>
> Yes, in general, the better the geometry is for efficiency and agility
> while walking, the less stable it will be when stationary. You may want to
> experiment to find the boundaries for static stability then go just far
> enough those to have a comfortable margin of safety.

Much like a fighter jet.  It needs to be unstable to react quickly.  I don't
need that fast a reaction time!  I'm opting for a little more stability.

> I can only guess here, but I presume that's why BigDog so often "prances"
> rather than stands still. It *can* stand still, because it rises from a
> camel crouch, but since it lacks feet (except in the beta version :), it's
> probably at risk of slipping when standing, even though all fours are on
> the ground.
>
> - Robin

I was thinking more along the lines that it's easier to make small
corrections while the machine is in motion.  Call it "dithering".

"Strider" (possible name for new robot) got his first exhibition last night
at the HBRobotics club meeting.  Managed to walk a little, and do a
pirouette, thanks to some 6:00 AM software touches.  More to do!

I noticed that the legs don't initialize exactly equal, even though I
thought I had "trimmed" them with PowerPod (LM program), and wrote to the
registers of the SSC32.  When the legs are not equally set up, the 'bot
veers off to one side quite easily.  I also moved the battery pack to the
extreme rear (with rubber bands!), which kept the rear in contact with the
floor/table better.  More experiments needed.

Visitors from Spain took pictures of strider and me, so maybe we'll end up
in their Spanish Tech newspaper!

Alan KM6VV

Hi Alan,

Strider is a cute name! It might need a sword to go with that ;)

Hey, it would be cool if you get a write up in the Spanish newspaper!!

- Robin




--- In HexapodRobotIK@yahoogroups.com, "Alan Marconett" <KM6VV@...> wrote:
> "Strider" (possible name for new robot) got his first exhibition last night
> at the HBRobotics club meeting.  Managed to walk a little, and do a
> pirouette, thanks to some 6:00 AM software touches.  More to do!
>
> I noticed that the legs don't initialize exactly equal, even though I
> thought I had "trimmed" them with PowerPod (LM program), and wrote to the
> registers of the SSC32.  When the legs are not equally set up, the 'bot
> veers off to one side quite easily.  I also moved the battery pack to the
> extreme rear (with rubber bands!), which kept the rear in contact with the
> floor/table better.  More experiments needed.
>
> Visitors from Spain took pictures of strider and me, so maybe we'll end up
> in their Spanish Tech newspaper!
>
> Alan KM6VV
>

Hi Robin,

Cute?  I was going for Rugged!  No matter!  Might just want to call it the
"Galloping Gertie"!

Yeah, you never know!  It wasn't too clear what they were collecting pix and
names for...

Alan KM6VV

> -----Original Message-----
> On Behalf Of Robin


> Strider is a cute name! It might need a sword to go with that ;)
>
> Hey, it would be cool if you get a write up in the Spanish newspaper!!
>
> - Robin

Oh, right. I forgot: Never use "cute" around a guy =))

Yes, definitely, Strider is a rugged name. Tough. (Esp. w/the sword :)

- Robin


--- In HexapodRobotIK@yahoogroups.com, "Alan Marconett" <KM6VV@...> wrote:
>
> Hi Robin,
>
> Cute?  I was going for Rugged!  No matter!  Might just want to call it the
> "Galloping Gertie"!
>
> Yeah, you never know!  It wasn't too clear what they were collecting pix and
> names for...
>
> Alan KM6VV
>
> > -----Original Message-----
> > On Behalf Of Robin
>
>
> > Strider is a cute name! It might need a sword to go with that ;)
> >
> > Hey, it would be cool if you get a write up in the Spanish newspaper!!
> >
> > - Robin
>

I need to work on the sword!

Here's a mention of the da Vinci Surgical robot:

http://www.forbes.com/2009/05/27/intuitive-surgical-davinci-business-robots.
html?partner=yahootix

Alan  KM6VV

> Oh, right. I forgot: Never use "cute" around a guy =))
>
> Yes, definitely, Strider is a rugged name. Tough. (Esp. w/the sword :)
>
> - Robin

Hey Alan,

Dang I've been so caught up in various things this past month that I didn't even
look at Servo Mag until yesterday. Congratulations on making it onto the cover!

- Robin

Thanks Robin!

It was a nice early birthday present.  Loki, my previous 2-part article also
made the cover last June-July.  I'm very pleased!

Alan KM6VV

> Hey Alan,
>
> Dang I've been so caught up in various things this past month that I
> didn't even look at Servo Mag until yesterday. Congratulations on making
> it onto the cover!
>
> - Robin
>
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>

Hey, that's great! And Happy Birthday =)

- Robin


--- In HexapodRobotIK@yahoogroups.com, "Alan Marconett" <KM6VV@...> wrote:
>
> Thanks Robin!
>
> It was a nice early birthday present.  Loki, my previous 2-part article also
> made the cover last June-July.  I'm very pleased!
>
> Alan KM6VV
>
> > Hey Alan,
> >
> > Dang I've been so caught up in various things this past month that I
> > didn't even look at Servo Mag until yesterday. Congratulations on making
> > it onto the cover!
> >
> > - Robin
> >
> >
> >
> >
> > ------------------------------------
> >
> > Yahoo! Groups Links
> >
> >
> >
>

anyone can give me tips how to control 16 servo at hexapod using atmega 8535?

Hi ?,

get an SSC-32 R/C servo driver from Lynxmotion!

Alan  KM6VV

ipammm_el06 wrote:
> anyone can give me tips how to control 16 servo at hexapod using atmega 8535?
>
>
>

Friend of mine sent me a link to a FOX new article:
http://www.foxnews.com/scitech/2010/02/09/spider-robot-learns-walk/?utm_source=f\
eedburner&utm_medium=feed&utm_campaign=Feed%253A+foxnews%252Fscitech+%2528Text+-\
+SciTech%2529

Did some searching and found an Intel tie in

      http://edc.intel.com/Applications/Industrial/Hexapod-Robot/

Thought i would share :-)

-donnie

Yes!

That's very much on point!  Thanks for posting!

Alan  KM6VV

> -----Original Message-----
> On Behalf Of Donnie Savage
>
> Friend of mine sent me a link to a FOX new article:
> http://www.foxnews.com/scitech/2010/02/09/spider-robot-learns-
> walk/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%253A+foxnews
> %252Fscitech+%2528Text+-+SciTech%2529
>
> Did some searching and found an Intel tie in
>
>      http://edc.intel.com/Applications/Industrial/Hexapod-Robot/
>
> Thought i would share :-)
>
> -donnie

Interesting to read.
They say
The Hexapod robot was designed at the Robotics and Neural Systems Laboratory
(RNSL) at the University of Arizona
However reading the pdf they seem to mean the system was designed but the rest?
How did they manage to burn out the Polu serial servo controller? Designed?
Not really a good example of Q learning. Only bits were Q learned.
Why is it that when everybody can see insects have a front and a back and sides
do they go and 'design' circularly symetric hexapod robots.
Is it an emergent behaviour of CAD+human systems?
It is probably why you see a lot of videos of hexapods spinning round on the
spot, economy of code.
But they don't seem to be advancing the theory and practice of useful walking
robots very much.
David

   ----- Original Message -----
   From: Alan Marconett
   To: HexapodRobotIK@yahoogroups.com
   Sent: Thursday, February 11, 2010 6:16 PM
   Subject: RE: [HexapodRobotIK] Q-Learning Hexapod



   Yes!

   That's very much on point! Thanks for posting!

   Alan KM6VV

   > -----Original Message-----
   > On Behalf Of Donnie Savage
   >
   > Friend of mine sent me a link to a FOX new article:
   > http://www.foxnews.com/scitech/2010/02/09/spider-robot-learns-
   > walk/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%253A+foxnews
   > %252Fscitech+%2528Text+-+SciTech%2529
   >
   > Did some searching and found an Intel tie in
   >
   > http://edc.intel.com/Applications/Industrial/Hexapod-Robot/
   >
   > Thought i would share :-)
   >
   > -donnie





[Non-text portions of this message have been removed]

If the 'bot has no head with sensors (hobby robot), then it's easy enough to
translate or move in any direction.  If you do have a "head", then unless it
has full 360 deg rotation to track that of the motion, it's necessary to
"follow your head".

I think the first hexapods were rectangular.  Circularly symmetric robots
can translate (better).  My current designs give some length back to the
'bot.

Didn't know Matt was using an Intel Atom in his 'bots.  Must be a different
Matt.

I'll bet he took a look or two at Micromagic Systems, or Lynxmotion.

Alan KM6VV


> -----Original Message-----
> On Behalf Of David Buckley

> Interesting to read.
> They say
> The Hexapod robot was designed at the Robotics and Neural Systems
> Laboratory (RNSL) at the University of Arizona
> However reading the pdf they seem to mean the system was designed but the
> rest?
> How did they manage to burn out the Polu serial servo controller?
> Designed?
> Not really a good example of Q learning. Only bits were Q learned.
> Why is it that when everybody can see insects have a front and a back and
> sides do they go and 'design' circularly symetric hexapod robots.
> Is it an emergent behaviour of CAD+human systems?
> It is probably why you see a lot of videos of hexapods spinning round on
> the spot, economy of code.
> But they don't seem to be advancing the theory and practice of useful
> walking robots very much.
> David
>

Things that are alive have a head because they do things to things and generally
that taxes the visual system and brain fully, if it didn't creatures would have
smaller brains. Their centre of attention is not their centre of mass but the
location where things are being done.
Circular robots are good at being totally self centered which makes them poor at
doing stuff to things.

This line of 'design' is a dead end rather like the monowheel motorcycles of the
early 1900s (and sadly still) sure the big wheel diameter means they can drive
along rougher roads BUT they don't turn corners too well, watch the accident
videos.

The Micromagic leg design (I am sure that Matt was the first) looks a good
spider leg design. Now what would be a good next step would be for the
Hex(Octo)pod to climb over things like a spider does.
Of course that would mean a complete change in philosophy because getting the
terrain map into the robot so it could do the IK would at the moment be
virtually impossible except with massive off board computing.
You will note that all these IK hexapods need a flat surface.
David


   ----- Original Message -----
   From: Alan Marconett
   To: HexapodRobotIK@yahoogroups.com
   Sent: Thursday, February 11, 2010 8:16 PM
   Subject: RE: [HexapodRobotIK] Q-Learning Hexapod



   If the 'bot has no head with sensors (hobby robot), then it's easy enough to
   translate or move in any direction. If you do have a "head", then unless it
   has full 360 deg rotation to track that of the motion, it's necessary to
   "follow your head".

   I think the first hexapods were rectangular. Circularly symmetric robots
   can translate (better). My current designs give some length back to the
   'bot.

   Didn't know Matt was using an Intel Atom in his 'bots. Must be a different
   Matt.

   I'll bet he took a look or two at Micromagic Systems, or Lynxmotion.

   Alan KM6VV

   > -----Original Message-----
   > On Behalf Of David Buckley

   > Interesting to read.
   > They say
   > The Hexapod robot was designed at the Robotics and Neural Systems
   > Laboratory (RNSL) at the University of Arizona
   > However reading the pdf they seem to mean the system was designed but the
   > rest?
   > How did they manage to burn out the Polu serial servo controller?
   > Designed?
   > Not really a good example of Q learning. Only bits were Q learned.
   > Why is it that when everybody can see insects have a front and a back and
   > sides do they go and 'design' circularly symetric hexapod robots.
   > Is it an emergent behaviour of CAD+human systems?
   > It is probably why you see a lot of videos of hexapods spinning round on
   > the spot, economy of code.
   > But they don't seem to be advancing the theory and practice of useful
   > walking robots very much.
   > David
   >





[Non-text portions of this message have been removed]

I think Matt is working to this end.  He can already negotiate minor
terrain.

But you're right.  All the other hexapods and octopods I know about are
still basically confined to level (at least smooth) surfaces.

I wonder if the 'Q' bot is actually relying on its onboard computer, or if
it's using some processing power off of the PC.  I noticed it has an
umbilical for power at least.

Interesting to note that the parts (legs?) were "printed".

Alan KM6VV

> -----Original Message-----
> On Behalf Of David Buckley
>
> Things that are alive have a head because they do things to things and
> generally that taxes the visual system and brain fully, if it didn't
> creatures would have smaller brains. Their centre of attention is not
> their centre of mass but the location where things are being done.
> Circular robots are good at being totally self centered which makes them
> poor at doing stuff to things.
>
> This line of 'design' is a dead end rather like the monowheel motorcycles
> of the early 1900s (and sadly still) sure the big wheel diameter means
> they can drive along rougher roads BUT they don't turn corners too well,
> watch the accident videos.
>
> The Micromagic leg design (I am sure that Matt was the first) looks a good
> spider leg design. Now what would be a good next step would be for the
> Hex(Octo)pod to climb over things like a spider does.
> Of course that would mean a complete change in philosophy because getting
> the terrain map into the robot so it could do the IK would at the moment
> be virtually impossible except with massive off board computing.
> You will note that all these IK hexapods need a flat surface.
> David
>

Hi Alan,

Iv been following this discussion as well as following this project from the
start. here is his link to the "Build".
Intel have now employed him i believe, or at least i know Intel now pay for his
parts.
http://hexapodrobot.com/forum/viewtopic.php?f=5&t=345



To: HexapodRobotIK@yahoogroups.com
From: KM6VV@...
Date: Thu, 11 Feb 2010 13:13:01 -0800
Subject: RE: [HexapodRobotIK] Q-Learning Hexapod




























       I think Matt is working to this end.  He can already negotiate minor

terrain.



But you're right.  All the other hexapods and octopods I know about are

still basically confined to level (at least smooth) surfaces.



I wonder if the 'Q' bot is actually relying on its onboard computer, or if

it's using some processing power off of the PC.  I noticed it has an

umbilical for power at least.



Interesting to note that the parts (legs?) were "printed".



Alan KM6VV



> -----Original Message-----

> On Behalf Of David Buckley

>

> Things that are alive have a head because they do things to things and

> generally that taxes the visual system and brain fully, if it didn't

> creatures would have smaller brains. Their centre of attention is not

> their centre of mass but the location where things are being done.

> Circular robots are good at being totally self centered which makes them

> poor at doing stuff to things.

>

> This line of 'design' is a dead end rather like the monowheel motorcycles

> of the early 1900s (and sadly still) sure the big wheel diameter means

> they can drive along rougher roads BUT they don't turn corners too well,

> watch the accident videos.

>

> The Micromagic leg design (I am sure that Matt was the first) looks a good

> spider leg design. Now what would be a good next step would be for the

> Hex(Octo)pod to climb over things like a spider does.

> Of course that would mean a complete change in philosophy because getting

> the terrain map into the robot so it could do the IK would at the moment

> be virtually impossible except with massive off board computing.

> You will note that all these IK hexapods need a flat surface.

> David

>



















[Non-text portions of this message have been removed]

Hi Jonny,

Yeah, looks like he knows about the LM and Micromagic websites!  Nice work.
Sweet deal!  I have some magazine articles to help pay for my parts.  I
think Intel has some deeper pockets!

Thanks for the link.

Alan  KM6VV

> -----Original Message-----
> On Behalf Of jonny poole
>
> Hi Alan,
>
> Iv been following this discussion as well as following this project from
> the start. here is his link to the "Build".
> Intel have now employed him i believe, or at least i know Intel now pay
> for his parts.

can someone tell me what the controller is
where to buy it?
thank you

yau

--- In HexapodRobotIK@yahoogroups.com, "Alan Marconett" <KM6VV@...> wrote:
>
> Hi Jonny,
>
> Yeah, looks like he knows about the LM and Micromagic websites!  Nice work.
> Sweet deal!  I have some magazine articles to help pay for my parts.  I
> think Intel has some deeper pockets!
>
> Thanks for the link.
>
> Alan  KM6VV
>
> > -----Original Message-----
> > On Behalf Of jonny poole
> >
> > Hi Alan,
> >
> > Iv been following this discussion as well as following this project from
> > the start. here is his link to the "Build".
> > Intel have now employed him i believe, or at least i know Intel now pay
> > for his parts.
>

Hi All - most of the postings seem rather old here, so I'm hoping that this
group is still active.

Just to introduce myself to the group - I'm currently working on a project to
make a 2DOF hexapod (3DOF in time) as a starting point - my own ground rules are
to avoid expense and ready manufactured solutions where I can. So I'm only just
discovering problems that many of you will have solved many times before - bu
they the learning's what it's all about.

I'm finding the manufacture of the legs to be particularly challenging, as It
breaks both of my rules to buy servo Brackets at £10 a time ! I'm resigned to
knocking up my own but it'll take time. Hopefully I'll be posting my results in
due course.

I've already completed a couple of small projects (photovore and collision
detecting bug) and have recently produced a 4X4 wheeled platform, that I can use
as a test bed for the sensory kit I'm hoping to mount on the hex in time.

Hi Andrew
Welcome to a very quiet group, it isn't that there is nobody home, it is just
that we are not actively discussing.
However that could change and I am sure the group can offer useful advice.
You don't have to use expensive brackets, see
http://www.davidbuckley.net/DB/Simplhex.htm
and
http://www.davidbuckley.net/DB/Hextor.htm

DAvid- robots42
WE are the music makers,
and we are the dreamers of dreams.


   ----- Original Message -----
   From: andrew.cusick@...
   To: HexapodRobotIK@yahoogroups.com
   Sent: Tuesday, March 08, 2011 6:53 PM
   Subject: [HexapodRobotIK] Hi out there



   Hi All - most of the postings seem rather old here, so I'm hoping that this
group is still active.

   Just to introduce myself to the group - I'm currently working on a project to
make a 2DOF hexapod (3DOF in time) as a starting point - my own ground rules are
to avoid expense and ready manufactured solutions where I can. So I'm only just
discovering problems that many of you will have solved many times before - bu
they the learning's what it's all about.

   I'm finding the manufacture of the legs to be particularly challenging, as It
breaks both of my rules to buy servo Brackets at £10 a time ! I'm resigned to
knocking up my own but it'll take time. Hopefully I'll be posting my results in
due course.

   I've already completed a couple of small projects (photovore and collision
detecting bug) and have recently produced a 4X4 wheeled platform, that I can use
as a test bed for the sensory kit I'm hoping to mount on the hex in time.





[Non-text portions of this message have been removed]

Hi Ali,



I didn't see your post on the list.  Not much being discussed here.  There
are examples in the files.  My 'bot uses custom C code, not really
share-able, outdated.



I'd check out the Phoenix code on the Lynx Motion forum:



http://www.lynxmotion.net/index.php



http://www.lynxmotion.net/viewtopic.php?f=8
<http://www.lynxmotion.net/viewtopic.php?f=8&t=4285> &t=4285



And look at the Hexapod diagrams in the Photos.



Ask questions!



Alan KM6VV



From: "Ali" <alicute@...>
Date: January 15, 2012 1:14:10 AM PST
Subject: [HexapodRobotIK] Q-Learning Hexapod

hi, now i'm still build a hexapod robot. I want my hexapod robot using IK
and Q-Learning. but I don't know, how to study IK & Q-Learning. Can I ask
your source code Hexapod IK & Q-Learning?







[Non-text portions of this message have been removed]

Ali
Have you ever done Q-learning?
If not best to learn how to that on something simpler
http://www.informit.com/articles/article.aspx?p=26423
David

   ----- Original Message -----
   From: Alan
   To: HexapodRobotIK@yahoogroups.com
   Cc: alicute@...
   Sent: Monday, January 16, 2012 8:18 PM
   Subject: [HexapodRobotIK] Q-Learning Hexapod



   Hi Ali,

   I didn't see your post on the list. Not much being discussed here. There
   are examples in the files. My 'bot uses custom C code, not really
   share-able, outdated.

   I'd check out the Phoenix code on the Lynx Motion forum:

   http://www.lynxmotion.net/index.php

   http://www.lynxmotion.net/viewtopic.php?f=8
   <http://www.lynxmotion.net/viewtopic.php?f=8&t=4285> &t=4285

   And look at the Hexapod diagrams in the Photos.

   Ask questions!

   Alan KM6VV

   From: "Ali" <alicute@...>
   Date: January 15, 2012 1:14:10 AM PST
   Subject: [HexapodRobotIK] Q-Learning Hexapod

   hi, now i'm still build a hexapod robot. I want my hexapod robot using IK
   and Q-Learning. but I don't know, how to study IK & Q-Learning. Can I ask
   your source code Hexapod IK & Q-Learning?

   [Non-text portions of this message have been removed]





[Non-text portions of this message have been removed]

Hi David,

Good thing you picked up on the Q-learning, I wasn't aware of it!

Sounds like that would be a few levels up from getting the IK running.

Alan KM6VV

> -----Original Message-----
> On Behalf Of David Buckley
>
> Ali
> Have you ever done Q-learning?
> If not best to learn how to that on something simpler
> http://www.informit.com/articles/article.aspx?p=26423
> David
>
>   ----- Original Message -----
>   From: Alan
>   To: HexapodRobotIK@yahoogroups.com
>   Cc: alicute@...
>   Sent: Monday, January 16, 2012 8:18 PM
>   Subject: [HexapodRobotIK] Q-Learning Hexapod
>
>
>
>   Hi Ali,
>
>   I didn't see your post on the list. Not much being discussed here. There
>   are examples in the files. My 'bot uses custom C code, not really
>   share-able, outdated.
>
>   I'd check out the Phoenix code on the Lynx Motion forum:
>
>   http://www.lynxmotion.net/index.php
>
>   http://www.lynxmotion.net/viewtopic.php?f=8
>   <http://www.lynxmotion.net/viewtopic.php?f=8&t=4285> &t=4285
>
>   And look at the Hexapod diagrams in the Photos.
>
>   Ask questions!
>
>   Alan KM6VV
>
>   From: "Ali" <alicute@...>
>   Date: January 15, 2012 1:14:10 AM PST
>   Subject: [HexapodRobotIK] Q-Learning Hexapod
>
>   hi, now i'm still build a hexapod robot. I want my hexapod robot using
IK
>   and Q-Learning. but I don't know, how to study IK & Q-Learning. Can I
ask
>   your source code Hexapod IK & Q-Learning?
>
>   [Non-text portions of this message have been removed]
>
>
>
>
>
> [Non-text portions of this message have been removed]
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>

Hi Alan
Right now I have NO IDEA how to start Q learning IK
It seems to me the search space would be enormous. Not quite like monkeys typing
Shakespeare but near.
David

   ----- Original Message -----
   From: Alan
   To: HexapodRobotIK@yahoogroups.com
   Sent: Tuesday, January 17, 2012 12:43 AM
   Subject: RE: [HexapodRobotIK] Q-Learning Hexapod



   Hi David,

   Good thing you picked up on the Q-learning, I wasn't aware of it!

   Sounds like that would be a few levels up from getting the IK running.

   Alan KM6VV

   > -----Original Message-----
   > On Behalf Of David Buckley
   >
   > Ali
   > Have you ever done Q-learning?
   > If not best to learn how to that on something simpler
   > http://www.informit.com/articles/article.aspx?p=26423
   > David
   >
   > ----- Original Message -----
   > From: Alan
   > To: HexapodRobotIK@yahoogroups.com
   > Cc: alicute@...
   > Sent: Monday, January 16, 2012 8:18 PM
   > Subject: [HexapodRobotIK] Q-Learning Hexapod
   >
   >
   >
   > Hi Ali,
   >
   > I didn't see your post on the list. Not much being discussed here. There
   > are examples in the files. My 'bot uses custom C code, not really
   > share-able, outdated.
   >
   > I'd check out the Phoenix code on the Lynx Motion forum:
   >
   > http://www.lynxmotion.net/index.php
   >
   > http://www.lynxmotion.net/viewtopic.php?f=8
   > <http://www.lynxmotion.net/viewtopic.php?f=8&t=4285> &t=4285
   >
   > And look at the Hexapod diagrams in the Photos.
   >
   > Ask questions!
   >
   > Alan KM6VV
   >
   > From: "Ali" <alicute@...>
   > Date: January 15, 2012 1:14:10 AM PST
   > Subject: [HexapodRobotIK] Q-Learning Hexapod
   >
   > hi, now i'm still build a hexapod robot. I want my hexapod robot using
   IK
   > and Q-Learning. but I don't know, how to study IK & Q-Learning. Can I
   ask
   > your source code Hexapod IK & Q-Learning?
   >
   > [Non-text portions of this message have been removed]
   >
   >
   >
   >
   >
   > [Non-text portions of this message have been removed]
   >
   >
   >
   > ------------------------------------
   >
   > Yahoo! Groups Links
   >
   >
   >




[Non-text portions of this message have been removed]

Hi David,

Now that I think about it, there were paper(s) on a neural net and
learning leg movements, but I probably couldn't find my copy now.  No
source code for it.

Alan KM6VV

On 1/16/2012 8:05 PM, David Buckley wrote:
> Hi Alan
> Right now I have NO IDEA how to start Q learning IK
> It seems to me the search space would be enormous. Not quite like monkeys
typing Shakespeare but near.
> David
>
>    ----- Original Message -----
>    From: Alan
>    To: HexapodRobotIK@yahoogroups.com
>    Sent: Tuesday, January 17, 2012 12:43 AM
>    Subject: RE: [HexapodRobotIK] Q-Learning Hexapod
>
>
>
>    Hi David,
>
>    Good thing you picked up on the Q-learning, I wasn't aware of it!
>
>    Sounds like that would be a few levels up from getting the IK running.
>
>    Alan KM6VV
>
>    >  -----Original Message-----
>    >  On Behalf Of David Buckley
>    >
>    >  Ali
>    >  Have you ever done Q-learning?
>    >  If not best to learn how to that on something simpler
>    >  http://www.informit.com/articles/article.aspx?p=26423
>    >  David
>    >
>    >  ----- Original Message -----
>    >  From: Alan
>    >  To: HexapodRobotIK@yahoogroups.com
>    >  Cc: alicute@...
>    >  Sent: Monday, January 16, 2012 8:18 PM
>    >  Subject: [HexapodRobotIK] Q-Learning Hexapod
>    >
>    >
>    >
>    >  Hi Ali,
>    >
>    >  I didn't see your post on the list. Not much being discussed here. There
>    >  are examples in the files. My 'bot uses custom C code, not really
>    >  share-able, outdated.
>    >
>    >  I'd check out the Phoenix code on the Lynx Motion forum:
>    >
>    >  http://www.lynxmotion.net/index.php
>    >
>    >  http://www.lynxmotion.net/viewtopic.php?f=8
>    >  <http://www.lynxmotion.net/viewtopic.php?f=8&t=4285>  &t=4285
>    >
>    >  And look at the Hexapod diagrams in the Photos.
>    >
>    >  Ask questions!
>    >
>    >  Alan KM6VV
>    >
>    >  From: "Ali"<alicute@...>
>    >  Date: January 15, 2012 1:14:10 AM PST
>    >  Subject: [HexapodRobotIK] Q-Learning Hexapod
>    >
>    >  hi, now i'm still build a hexapod robot. I want my hexapod robot using
>    IK
>    >  and Q-Learning. but I don't know, how to study IK&  Q-Learning. Can I
>    ask
>    >  your source code Hexapod IK&  Q-Learning?
>    >
>    >  [Non-text portions of this message have been removed]
>    >
>    >
>    >
>    >
>    >
>    >  [Non-text portions of this message have been removed]
>    >
>    >
>    >
>    >  ------------------------------------
>    >
>    >  Yahoo! Groups Links
>    >
>    >
>    >
>
>
>
>
> [Non-text portions of this message have been removed]
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>