highaltitude.log.20081120

[00:04] Ebola (n=Ebola@unaffiliated/ebola) left irc: "Terry you slag! You knicked the leg of time!"
[00:05] rjharrison (n=rharriso@80.176.172.227) left #highaltitude.
[03:37] Laurenceb (i=83e3dd6d@gateway/web/ajax/mibbit.com/x-5df9af91e7caaf1c) left irc: "http://www.mibbit.com ajax IRC Client"
[03:55] jiffe88 (n=jiffe2@67.158.3.217) joined #highaltitude.
[07:41] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) joined #highaltitude.
[07:57] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) left irc:
[07:58] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) joined #highaltitude.
[07:58] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) left irc: Client Quit
[08:13] Simon-MPFH (n=simon@phantom.mpfh.co.uk) joined #highaltitude.
[10:11] Laurenceb (i=83e34f19@gateway/web/ajax/mibbit.com/x-64cf7975a967009d) joined #highaltitude.
[10:14] Ebola (n=Ebola@unaffiliated/ebola) joined #highaltitude.
[11:24] Simon-MPFH (n=simon@phantom.mpfh.co.uk) left irc: Read error: 131 (Connection reset by peer)
[11:24] Simon-MPFH (n=simon@phantom.mpfh.co.uk) joined #highaltitude.
[11:31] Ebola (n=Ebola@unaffiliated/ebola) left irc: "time to go make some tablets"
[14:32] natrium42 (n=alexei@CPE000625d867e2-CM0014045885be.cpe.net.cable.rogers.com) left irc:
[15:22] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) joined #highaltitude.
[15:28] edmoore (n=edmoore@pomegranate.chu.cam.ac.uk) left irc:
[16:21] <EI5GTB> evening
[16:27] <Laurenceb> hi
[16:36] <EI5GTB> how goes it?>
[16:36] <EI5GTB> http://tom.pycke.be/mav/71/kalman-filtering-of-imu-data
[16:36] <EI5GTB> makes it a bit simpleer
[16:36] natrium42 (n=alexei@24.114.233.17) joined #highaltitude.
[16:45] <Laurenceb> thats a big fudge
[16:45] <Laurenceb> its not a full state estimator
[16:49] natrium (n=alexei@24.114.232.18) joined #highaltitude.
[16:55] natrium (n=alexei@24.114.232.18) left irc: Remote closed the connection
[17:01] natrium (n=alexei@24.114.232.18) joined #highaltitude.
[17:02] natrium42 (n=alexei@24.114.233.17) left irc: Read error: 110 (Connection timed out)
[17:05] <Laurenceb> http://imagebin.org/31704
[17:05] <Laurenceb> thats what I've got with the latest script
[17:06] <Laurenceb> basically if you imagine blue is a point on the top and red is say one of the wingtips
[17:07] <Laurenceb> thats using some data recorded using the sparkfun 6DOF IMU - a guy on the forum recorded it
[17:12] natrium42 (n=alexei@24.114.233.18) joined #highaltitude.
[17:20] natrium42 (n=alexei@24.114.233.18) left irc: Read error: 60 (Operation timed out)
[17:33] natrium (n=alexei@24.114.232.18) left irc: Read error: 110 (Connection timed out)
[18:02] Hiena (n=Hiena@81.93.195.181.datatrans.hu) joined #highaltitude.
[18:16] Ebola (n=Ebola@unaffiliated/ebola) joined #highaltitude.
[18:27] <EI5GTB> Laurenceb, i now understand the concept of filtering the actual data using the kalman algoritim... but relating the 3 axis together....
[18:39] <Laurenceb> hehe
[18:41] <EI5GTB> lol
[18:41] <EI5GTB> maybe a 17 year old just isnt supposed to understand..
[18:43] <Laurenceb> http://pastebin.com/m5185c638
[18:43] <Laurenceb> thats my script
[18:44] <EI5GTB> hmm
[18:44] Action: EI5GTB has matlab downloading atm
[18:44] <EI5GTB> http://academic.csuohio.edu/simond/courses/eec644/kalman.pdf
[18:45] <EI5GTB> i read that, explains it fairly well
[18:45] <EI5GTB> i just start to loose track of what each letter stands for =/
[18:50] <EI5GTB> so.. the accelerometers are... keeping the gyros in check? and they are defined as a function of... prsuming where they are in relation to the gyros..
[18:50] <EI5GTB> ...
[18:50] <Laurenceb> yeah
[18:50] <Laurenceb> they keep drift under control
[18:52] <EI5GTB> aha.. ok
[18:52] <EI5GTB> now its starting to become clear
[18:52] Ebola (n=Ebola@unaffiliated/ebola) left irc: Remote closed the connection
[18:53] <EI5GTB> so that takes care of the inter sensor interactio on the single axis....... but... are the xy and z axies interrelated?
[18:56] <Laurenceb> yes
[18:56] <Laurenceb> it gets compicated
[18:57] <EI5GTB> grr
[18:58] <EI5GTB> also, with the thing actually moving around the air.... and accelerating... will that effect the acceleromters input to the calman filter?
[18:58] <Laurenceb> yes
[18:58] <EI5GTB> i.e is the thing modeld on the thing only pitching around one fixed point
[18:58] <EI5GTB> ah
[18:58] <EI5GTB> hmm, it all gets very messy then..
[18:59] <Laurenceb> I think a basic 3 state filter on an AVR would be enough to run a quadcopter
[18:59] <Laurenceb> i.e. the one I got working yesterday
[18:59] <EI5GTB> yea
[18:59] <Laurenceb> I've moved up to an 8 state filter now :P
[18:59] <EI5GTB> i think i understand states now... is it like the attitude one second ago, the attitude now and the attitude one second from now?
[19:00] <Laurenceb> yeah
[19:00] <Laurenceb> apart from its running much faster
[19:00] <Laurenceb> ~50 Hz
[19:00] <EI5GTB> i see
[19:01] <EI5GTB> so a 6 state would be [k-2][k-2][k][k+1][k+2]
[19:01] <EI5GTB> wait, thats 5
[19:01] <EI5GTB> exept its running slower... so k is... smaller...
[19:02] <EI5GTB> therefore each state has less of an effect on the outcome? or X as i think its refered to
[19:02] <Laurenceb> no
[19:03] <Laurenceb> number of "states" is probably a bad way to describe it
[19:03] <Laurenceb> 6 -> number of components oif the state vector
[19:03] <Laurenceb> so k is a 6 component (or 5) vector
[19:04] <EI5GTB> component... hmm
[19:04] Action: EI5GTB looks it up
[19:08] <EI5GTB> the components are somthing to do with the nmatricies?
[19:09] <Laurenceb> no
[19:09] <Laurenceb> (1,2,3,4)
[19:09] <Laurenceb> has 4 components
[19:18] Laurenceb (i=83e34f19@gateway/web/ajax/mibbit.com/x-64cf7975a967009d) left irc: "http://www.mibbit.com ajax IRC Client"
[19:43] natrium42 (n=alexei@72.136.18.113) joined #highaltitude.
[19:44] <natrium42> hi people and naxxfish
[19:45] <EI5GTB> hai
[19:48] <natrium42> hi paul
[20:13] Laurenceb (i=83e3dd6d@gateway/web/ajax/mibbit.com/x-851ff57c807f6b8b) joined #highaltitude.
[20:30] <EI5GTB> natrium42, how goes it?
[20:30] <EI5GTB> sorry, a bit distracted here
[20:44] Hiena (n=Hiena@81.93.195.181.datatrans.hu) left irc: "-=Halt! Hammerzeit!=-"
[21:19] rjharrison (n=rharriso@80.176.172.227) joined #highaltitude.
[21:19] <rjharrison> evening all
[21:19] <rjharrison> UK satellite coming over in 3 mins
[21:22] Ei5GTB_ (n=Paul@78.16.225.144) joined #highaltitude.
[21:22] <Ei5GTB_> dam isp
[21:26] EI5GTB (n=Paul@213-202-137-207.bas502.dsl.esat.net) left irc: Read error: 60 (Operation timed out)
[21:26] Nick change: Ei5GTB_ -> EI5GTB
[21:40] rjharrison (n=rharriso@80.176.172.227) left #highaltitude.
[21:55] fnoble (n=fnoblef@fn217.quns.cam.ac.uk) left irc: Read error: 110 (Connection timed out)
[21:57] Simon-MPFH (n=simon@phantom.mpfh.co.uk) left irc: "Leaving"
[22:02] edmoore (n=edmoore@chu-gw.churchillcambridge.co.uk) joined #highaltitude.
[22:15] Ebola (n=Ebola@unaffiliated/ebola) joined #highaltitude.
[22:51] <Laurenceb> hi edmoore
[23:04] <Laurenceb> edmoore:I've got some better kalman filtering working
[23:05] <edmoore> Laurenceb: fergus here, ed'll be back in a few mins
[23:05] <Laurenceb> hi fergus :P
[23:06] <Laurenceb> using an 8 component state vector
[23:08] <Laurenceb> I was thinking about the lassen iq you could use the ephemeris, dopplers and pseudoranges to input into position and velocity solution
[23:08] <Laurenceb> whats the clock speed on your arm?
[23:08] edmoore (n=edmoore@chu-gw.churchillcambridge.co.uk) left irc:
[23:09] <Laurenceb> bah
[23:14] edmoore (n=edmoore@chu-gw.churchillcambridge.co.uk) joined #highaltitude.
[23:14] <Laurenceb> wb
[23:14] <Laurenceb> I was thinking about the lassen iq you could use the ephemeris, dopplers and pseudoranges to input into position and velocity solution
[23:14] <Laurenceb> whats the clock speed on your arm?
[23:23] <Laurenceb> edmoore?
[23:23] <natrium42> Laurenceb, whatever you like it to be :)
[23:23] <Laurenceb> I was thinking about badger
[23:25] <Laurenceb> guess you can set it with the PLL
[23:27] <Laurenceb> rofl http://www.theregister.co.uk/2008/11/20/windows_vista_overtime/
[23:29] <natrium42> wtf, vista boots faster than xp
[23:36] Ebola (n=Ebola@unaffiliated/ebola) left irc: "sleep!"
[00:00] --- Fri Nov 21 2008