Sources of noise in geostationary satellite attitude determination simulation

[shakeel001]

I am doing satellite attitude determination simulation using sun Earth sensors and Gyro modelling .I am using extended kalman filtering for attitude determination.

the sequence of operations I am doing are
% Simulation data
% Spacecraft data
% Gyro data
% Sensor data.
% Control system initialization
% Generate the orbit
% el = [a,i,W,w,e,M]. The spacecraft is in geostationary orbit
% Initial conditions at equinox
% Sun and Earth vectors
% Run the simulation

I need to know the sources of error in determination of satellite attitude.How can I compare that the estimated value is correct.why the gyro is noisy always and what are the sources of errors which shall results to get noisy readings of sensors or other actuators.
I shall be very grateful to all for help in this regards

Thanks again in advance for your time and efforts

 

[D H]

There are several error sources. Some are

• Alignment error: The gyro axes are not pointing where you think they are. The IMU case might be misaligned, and the gyros might be misaligned inside the case.
  
• Non-orthogonality error: The gyros axes are not truly orthogonal to one another.
  
• Bias: A non-rotating object will get a non-zero reading. Making matters worse, the bias tends to drift over time.
  
• Scale factor error: The multiplicative factor used to convert counts to engineering units is not quite right. Like bias, scale factor error isn't quite constant. It can vary with temperature, how old the gyro is, and other factors.
  
• Random walk error: Even if everything else goes right, a gyro still has the basic problem of measuring a noisy signal. Assuming this is white, the output of a gyro is integrated white noise: a random walk.

Gyro spec sheets specify limits to many of the above; almost all have a random walk spec, sometimes called Allan Variance.

 

[shakeel001]

an you quote one example of each

1. Scale factor error which is temperature dependent

2. Random walk error

what are the sources of getting noise free measurements having sun,star and gyro or any other sensor which can provide noise free measurements in a satellite.

Thanks you very much and Regards

 

[D H]

There is no such thing as a noise-free measurement. Every sensor has errors. Part of your job is to properly characterize these errors. As for the source of the measurement, that's your job too. Your simulation needs to provide the truth data to the sensor model.

Writing a simulation engine has been the subject of many aerospace engineers masters theses. Using some pre-built simulation engine is typically an upper level undergraduate / lower level graduate aerospace engineering class. Spacecraft sensors, and modeling them, is often yet another class (or multiple classes; that topic is open-ended).

In other words, you are asking for (demanding in a big font!) a synopsis of multiple advanced classes. You might want to rethink your position here.