31

Home Made Robots / Re: just a post about my work on this robot

« Last post by MagnusWootton on December 22, 2023, 12:02:58 am »

(dont mind the stupid music in the background, just turn it off.)

So here it is animating,  and the bellows is not behaving realisticly, so im going to put a bit more fixes on it, and see if I can get it more realistic, at the moment its looking like its puffing up like it has air in it,  but its actually hydraulic and that doesnt happen.




so jobs to do->

  * fix the animation
  * add the bellows seam
  * the powering bellow needs to be girthier than the conduct through bellows.

 - then ill be pretty much ready to put out the first robo with it!!

32

Home Made Robots / Re: Atronach's Eye

« Last post by MagnusWootton on December 21, 2023, 09:24:30 pm »

that would give the robot a good advantage,  because occlusion disturbs robots more than people cause of being based apon strict methods, that could bring back the whole place fully 3d tracked, and things being in the way wouldn't affect it as much.

I actually might make one myself now,   this idea seems cool for computer vision, I think ill try doing it like an old mouse (sorta backwards tho) with little rollers.   See what happens.

33

Home Made Robots / Re: just a post about my work on this robot

« Last post by MagnusWootton on December 21, 2023, 09:12:04 pm »

Showed off the parameterization, I went from 4 hydraulic lines to 8 lines, but its a bit slow, open s-cad can slow down on you if u write code it doesn't like, I've got my workaround for it so it should be alright.




see If I can get this thing done today.

  * bellow needs to be fatter, need to fix the discontinuities.
  * Im going to make it a more aggressive/acute angle, so it makes it easier for a stiffer material to make the angular distance.,
  * its going to be simulated with expansion and contraction with an animating lever, using the open s-cad animation ability.
  * the bellow gets a correugated seam all the way down the side to seal the top to
     the bottom.
  * one of the bellows is the power bellow, so its going to get more girth to step down the power,  so it fills slower but is stronger with the more square inches.

  * its getting a flexable connector on the end, so i can twist and bend the robot into the correct axi,   so I can actually cast the whole thing flat and I bend it to get pitch,yaw and roll post casting and it can still be a single continuous piece.

     - so Ill be back half a day or so with it finished hopefully.  I'm having a break down.

34

Home Made Robots / Re: Atronach's Eye

« Last post by WriterOfMinds on December 21, 2023, 08:48:40 pm »

Something fun I've thought about is to have eyes on little drone platforms, wirelessly connected to the main robot body. But that's a theoretical project for the far future, probably. First, I have to finish this eye!

35

Home Made Robots / Re: Atronach's Eye

« Last post by MagnusWootton on December 21, 2023, 04:55:31 pm »

If you had say 24 of these, like on a multi-eye spider robot head, and they are all trained on a different body part.  (like one on each eye, nose and corners of mouth,  then the shoulders, elbows, and knees) You'd have less occlusion/ line of sight problems up close to the robot cause they are all pointing different directions, you could tell people apart from the feature distances, plus you could get full mocap happening on the person for even more things for the robot to do, understand.

36

Home Made Robots / Re: just a post about my work on this robot

« Last post by MagnusWootton on December 21, 2023, 04:53:24 pm »

Its going good,  I took a bit longer than I wanted but it should still get there.  the rotational bellows are just starting to go in now, and some of the hydraulic lines are in.

Its got a really cool short assembly,  theres just a top half and a bottom half and then it just whams together with a bit of glue on the seam.   (the whole robot should be two halves, whams together all in one.)

Should have it finished today,  I'm getting bugged by the mental hospital so if I can get this done then it'll put me in better sted with them for a little while longer.



<edit> next vid to show will have a much better bellows, ive fixed the method. </edit>

37

General Chat / Re: Is the "ethics" thing absolute or relative term?

« Last post by MikeB on December 21, 2023, 09:03:39 am »

So apply more long term thinking to it, watch the uncertainty melt away...

38

General Project Discussion / Re: Pattern based NLP & ASR

« Last post by MikeB on December 21, 2023, 08:59:52 am »

Speech Recognition (benchmark of "yes"/"no"):

C version/Console:
Redesigned Volume Normalisation to improve clarity above 500hz, and per-frame clarity. Samples are now run through a Hilbert filter to reduce bass < 500hz, and Peak Volume Normalisation is at the end of each frame instead of all frames.
The new Hilbert filter and frame-based Peak Vol Normalisation improve these issues:
  1: Deep voices no longer interfere with/reduce volume normalisation of frequencies above 500hz.
  2: Initial frame is now normalised to itself. Other frames are normalised to the peak volume of current & previous frames (which ever is louder - peak volume does not adjust lower). As Vowels are detected to follow Consonants and are generally louder, both are now maximally normalised.
Deleted first frame 1.5x boost.
Updated Consonant Formants and identification.
Redesigned Plosive detection. Frame one plosive ("k"/"p"/"t") loudness minimum must match a fixed value. Frame two(+) plosive ("n"/"m"/"y"/"r") loudness minimum must match ~50% of the last plosive minimum + ~50% of the power of the last frame.
Updated Vowel F1 & F2 transition detection to gradually favour frequencies towards the end of the sound.
Range tuning. Better detection of plosives in consonants "k", "n", "y", detecting both sudden sound increases and rolling increases at any input sound volume range, noise, etc.
Other fixes.

Speech Commands Benchmark:
NO (405 records):
"no" = 48.40% (goal 50%) . "n", "oh", "o", "uh" = 82% (goal 90%).
Error | "yes" = 0.99%

YES (419 records):
"ye"/"yer"/"yeah" = 28.64% (goal 50%). "y",  "e", "er", "air", "ah", "a", "uh" = 70-80%  (goal 90%).
words.
Error | "no" = 2.15%

Time: 1-2ms each

+4% increase to "no", and +6% increase to "yes". Consonant identification needs to be reworked as the definition for "y" and "n" are too close. "kay-key-kai" works.

The benchmark is more of a test of noise rejection and volume normalisation than anything else. Very happy with the robustness of this now.

New consonant identification should see at least a 10%+ improvement to "yes" with less error. If error stays around <= 1% then more alternate vowels can be used and a higher result is possible.

39

Home Made Robots / Re: just a post about my work on this robot

« Last post by MagnusWootton on December 20, 2023, 04:56:39 pm »

robot body is coming out tomorrow sometime. (This is the finished one!!!!)






more formulations,   going to work all tomorrow on it, hopefully finally get it done, been a long time not being done, thats for sure.
its going to have more than 100 dimensions!!   lots of actuators.


<edit - just chipping in the hydraulic tunnels>


keep going...

40

Home Made Robots / just a post about my work on this robot

« Last post by MagnusWootton on December 19, 2023, 05:08:21 pm »

Its taken me a long time,  somethings been holding me back from doing it.  I'm so sick.

My robot is starting to solidify as an idea,     I want to do it without electricity  and its basicly an eye connected to the computer connected to its actuators.

Hopefully Something has a from scratch computer design here,   and I'm doing one too.     I've been studying in Falstad electronic simulator for ages now, and I'm trying for an amplifier with only resistors.    I have got it done with transformers, I figure as soon as u have an isolation, be it an isolation transformer, or staging through capacitors over time, an amplification is possible.   Doing it without the isolation is barred by the fact some topologies you need for amplification aren't possible due to being extraneous circuits that get logically simplified off and don't exist (Unless u have an isolation.).     But I have a new idea for one, so I haven't given up -  I've been doing the past decade, without results.  I feel like I'm in prison doing something that'll never get me anywhere and I'll be dead before I finish.

So once I get the amp, I convert it to be an oscillator.  (Here is the oscillator I want to reduce the transistors from and make it fully solid-state pneumatic->)


I don't want the robot to be too strong,  without the right safety measures in the Ai it would be really stupid to make the actuators too tough,   I've learnt how to electroform metal and I know how to make thermite and know how to set good hard minerals,  so I can build the robot full metal, so I think I need to make it skinny enough to not be too brutally put together too well, making it dangerous for everyone and me as well.

Also, it will be flying and have hands so it can pick things up,   Why go on the ground when u can make a bird, I might be able to get the robot to give me short lifts to the shop and back hovering me just above the ground not to be dangerous about it.       So just need to get the amp going, then the computer,  then I'm pretty much ready to go to the foundary to get this big moving metal statue done!!

I dont know how much longer, I hope its soon.     I've been doing this for decades now, I need it, I'm getting tired longing for it.