Replies to This Discussion

Permalink Reply by Aristotle on September 12, 2008 at 1:01pm

wow I wish I had read the other responses before I put mine in ;)

• ▶ Reply

Permalink Reply by Earl on March 20, 2009 at 7:44am

We could have one discussion area that is JUST for posting code. Each should have a BRIEF description of what it is and on what platform.

• ▶ Reply

Permalink Reply by Matt Fisher on March 20, 2009 at 11:05am

I think you should just get a pastebin setup running. Easy as pie.
http://pastebin.com/

• ▶ Reply

Permalink Reply by NorthSweden on March 20, 2009 at 1:39pm

Good points,

And i think a bulletin forum would be good to have to. much better forum functions and then we could have some good threads to post in.

better control, but this Ning thing is good but limited.

• ▶ Reply

Permalink Reply by Sesan FESOBI on March 23, 2009 at 6:25pm

Hi everyone,
I have this nagging problem I've been trying to solve for the past 6 weeks. And that is how to simulate the optimisation of the integrated GPS/INS system with a Kalman using MATLAB. That is to demonstrate that the Kalman filter bounds the individual errors of the sensors (ie GPS and INS). Can anyone deliver me from this nightmare please!!!!

• ▶ Reply

Permalink Reply by Camille on March 23, 2009 at 10:26pm

Hey,
something like that?

% Simulation time limits------------------------
imax=length(time1);
jmax=length(time3);
tmax=Te*imax;

% Sensor errors---------------------------------


Acorrx=0.21472;
Acorry=0.50494;
Acorrz=0.009923;

Mcorrx=-0.021394;
Mcorry=0.019784;
Mcorrz=-0.03047;

% Initial magnetic and gravity reference vectors--
ref = 'avion';
% ref = 'model';
Init_references

% Initial state and covariances--------------------
Init_state
Init_covariances

%Initialisation of Kalman variables--------------
x=x0;
P=P0;
K=zeros(10,1);

%Temporal vectors and arrays for plotting----------
tt=[];
xx=[];
PP=[];
KK=[];

i=1; %index for acc and gyro updates
j=1; %index for magnetic updates
k=1; %index for gps updates

% Simulation----------------------------------------
for i=1:imax, %index for acc and gyro updates

t=(i-1)*Te;
[x,P]=prediction(x,P,Q,Te);
tt=[tt t];

% Angular rate measurements--------------------
pm=(Gyrox(i))*pi/180;
qm=(Gyroy(i))*pi/180;
rm=(Gyroz(i))*pi/180;
[x,P,K]=correction(x,P,R, 5,pm);
[x,P,K]=correction(x,P,R, 6,qm);
[x,P,K]=correction(x,P,R, 7,rm);

% Gravity measurements----------------------------
ax=Accx(i)-Acorrx; % Gravity measurements in body frame
ay=Accy(i)-Acorry;
az=Accz(i)-Acorrz;
[x,P,K]=correction(x,P,R,11,ax);
[x,P,K]=correction(x,P,R,12,ay);
[x,P,K]=correction(x,P,R,13,az);

% Magnetic measurements-------------------------
if j<=length(time3)
t_magn=time3(j)/1000;
if t_magn<=t+Te/2 & t_magn>=t-Te/2 % Nearest time coincidence into Te grid
mx=Magx(j)-Mcorrx; % Magnetic measurements in body frame
my=Magy(j)-Mcorry;
mz=Magz(j)-Mcorrz;
M=unity([mx my mz]');
mx=M(1);
my=M(2);
mz=M(3);
[x,P,K]=correction(x,P,R, 8,mx);
[x,P,K]=correction(x,P,R, 9,my);
[x,P,K]=correction(x,P,R,10,mz);
j=j+1;
end
end

xx=[xx x];
PP=[PP diag(P)];
KK=[KK K];

end

% Figures---------------------------------------------
% Raw_plots
Time_plots
Evol_plot



I hate nightmares so...

email me ;)

• ▶ Reply

• ‹ Previous
• 1
• 2
• 3
• Next ›
• Page