Answering only the first question: Yes, it is a realistic way.
Precision may be questionable however (so, she will figure out she is a few hundred million years ahead in time, but won't be sure whether it's 500M or 600M, if we presume present-day understanding of CMB, universe expansion and all the relevant stuff, and Planck-level precision of measurement. With future tech, she certainly can have precision under a million years.
So, you want to measure CMB. It would be useless to observe its patterns, as you won't know what those are in your new position, and those change on the scale of 100 thousand years anyway. So the only thing you can use is CMB temperature. Fortunately for you, we know how that behaves.
$$
T_{CMB}(t)=\frac{T_{CMB}(0)}{a(t)},
$$
where $T_{CMB}(0)$ is the present-day temperature,
$T_{CMB}(t)$ is temperature you want to measure, and $a(t)$ is the scale factor (defined to be 1 at the present day).
To properly describe how $a$ changes over time you would need to integrate Friedmann equation. Which is something your hero would do, but I'm too lazy for that. Fortunately, there is a good enough proportionality: in the current dark-energy-dominated era
$$
a(t) \propto exp(Ht),
$$
where $H$ is the Hubble constant. Plugging in $H=70km*s^{-1}*Mpc^{-1}$ and $t=570My$, we get
$$
a(t) \approx 1.042
$$
That means, as the first approximation, CMB temperature would drop by about 4%, or 0.11K. That's certainly a noticeable and measurable change even with present-day (if state of the art) detectors.
The problem with precision in our day arises from the uncertainty about Hubble constant. For example, we have two values ($67.66 m*s^{-1}*Mpc^{-1}$ and $74.03 m*s^{-1}*Mpc^{-1}$) which are supposed to have precision of less than 2%, but you can see that they differ way more. So far this discrepancy wasn't resolved. There is also the issue of Hubble constant not being really constant. We know it changes over time, we know it changes not by much on smaller time scale, but we don't really know how it would change. All this factored in, you can have precision of about $\pm100My$ or so with current data. On the other hand, it is certainly not a stretch to say that even in 10 years, this would be improved to $\pm10My$, and with whatever future technology and science you have, precision can be plausibly at least a couple orders better.