In Chapter 3, you state multiple times that a product of two normally distributed random variables is normally distributed after normalization. This is confusing. A product of two PDFs of two normal distributions is a PDF of a normal distribution after normalization. The product of two normally dist...

Say we get a reading of door, and suppose that testing shows that the sensor is 3 times more likely to be right than wrong. We should scale the probability distribution by 3 where there is a door. If the reading is door and the sensor is 3 times more likely to be right than wrong, then P(door) shoul...

The following comment on the homepage of the documentation is out of date: It’s already a bit hard to read because of the dot function calls (required because Python does not yet support an operator for matrix multiplication). This relates to: K = dot(P, H.T).dot(inv(dot(H, P).dot(H.T) + R)) Sinc...

Hello Dear Roger, I am studding your wonderful book on KF and thank you from all my heart. In the third chapter you have an error in the formula of a product of random variables, see the correct one on Wiki page: https://en.wikipedia.org/wiki/Distribution_of_the_product_of_two_random_variables Good...

I get an error during kalman_filter.update call with numpy 1.25., but if I roll back to numpy 1.24., this error does not occur. Thank you **_Python 3.11 Numpy: 1.25 Filterpy: 1.4.5 Traceback (most recent call last): File "/builds/model/kalman_filter_model.py", line 297, in update self.kalman_filter....

Thank you, sir, I am currently learning Kalman filtering, there are many formulas that I cannot understand, hope to have some examples to help me, just happened to see your work. I am very touched by your contribution to knowledge, haha, I also have no money to buy MATLAB, and finally, I wish you ca...

I have downloaded the whole repo, unzipped and uploaded it to my google drive from where I am opening the jupyter notebooks. In this chapter, import book_format runs into an error. I have installed filterpy in a previous cell by the way. On the other hand, when I open the repo in binder I get no s...

It should be a first-order filter for a robot moving in 2D not a second.

"P = P¯ − KPzKT" -> "P = P − KPzKT" ->

"The cross variance is computed as:..." Is the word "variance" intended here or the "covariance" word should be used, like in the "10.5.2 Update Step" when the covariance is computed between the state and the measurements: "To compute the Kalman gain we first compute the cross covariance of the stat...