search

thomas-haslwanter / scikit-kinematics

Python functions for working with 3D kinematics.
Other
129 stars 45 forks source link

the Kalman implementation questions #31

Open yywe opened 5 years ago

yywe commented 5 years ago

according to my understanding, not quite understand the implementation of the Kalman: (1) ... z_k_pre = z_k.copy() ... //# Update state vector x_k x_k += np.array( K_k.dot(z_k-z_k_pre) ).ravel() here z_k - z_k_pre means the current observation subtract the previous observation. according to standard Kalman text book, shouldn't here be observation subtract prediction ? and why the prediction part is just simply set to the previous observation (z_k_pre = z_k.copy()) (2) .... x_k[:3] = np.zeros(3) x_k[:3] += tstep (-x_k[:3]+z_k[:3]) .... here the first 3 elements is forced to 0. so actually it is x_k[:3] += tstep (z_k[:3]), since z_k[:3] is the angular velocity, the result is the amount of changed angles.

any help?

thomas-haslwanter commented 5 years ago

I am just looking into that again, and I am seriously wondering if my implementation was correct. Currently, I cannot even find the definition of the QUEST that I used at the time :( Sorry - Will keep you updated.

gcosne commented 4 years ago

Hi @thomas-haslwanter,

I can see that the Kalman code have been changed since @stweiyy comment. Do you have any update on the definition of QUEST you used ? How confident are you in the Kalman implementation?

thank you for your help