Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I'm working through a pair of papers on Simultaneous Localization and Mapping and I'm having trouble with some of the notation as I lack some formal math education.

The papers can be found
and here

$P_{k|k}$ and $\Sigma_{t}^{-1}$ are both covariance matrices.

What does the $_{k|k}$ mean? I recall from probability, that P(y|x) means the probability of y given x, but that doesn't seem to make sense here.

With $\Sigma_{t}^{-1}$ I thought that $\sum$ is usually used for summation (and I initially confused it with summation!) Is there any significance to $\sigma$ being used to represent the covariance matrix or is it just historical accident?

Some other questions raised while formulating this question (answered in the comments below):

Later, on page 6, there is a formula $P(x_{2} | x_1^{(i)})$. Here I don't know what the superscript $^{(i)}$ represents.

Answer: $x_1^{(i)} \sim P(x_1)$ is the ith sample from the distribution $P(x_1)$

The information vector $b_{t} = \mu_{t}^{T}H_{t}$ looks like the mean times the information matrix, but I thought the mean was a scalar value, so I don't understand the Transpose symbol.

Answer: $\mu_{t}$ is a vector because it is the mean of the $\zeta$

Thanks for everyone who's helped me refine these questions!

share|cite|improve this question
+1 for remembering to mention the paper. – J. M. Dec 6 '11 at 1:34
@t.b. I was sent here from MO as it's a more appropriate forum. The MO post is now deleted. – munk Dec 6 '11 at 2:59
Okay, I removed the crosspost notification. Here's the still relevant part of BR's comment on MO: "if you take another look at the papers, you will see that the notations are clearly defined ($P_{k∣k}$ and $\Sigma_t$ are covariance matrices and $x^{(i)}_1$ is the $i$-th sample from a distribution)." – t.b. Dec 6 '11 at 3:08
In the paper of Thrun et al., $\mu_t$ is a vector (because $\xi_t$ is a vector). – Joel Reyes Noche Dec 6 '11 at 3:12
I would imagine that using $\Sigma$ for the covariance matrix came from the use of $\sigma$ for standard deviation (according to an interesting website I had never seen before, $\sigma$ was the original notation for standard deviation). If you are really curious about this, you could ask it as a separate question (perhaps using the history tag). – B R Dec 6 '11 at 5:52

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.