Extensive Northern Hemisphere glaciation started about 3.5 million years
ago (beginning the current Ice Age), and extensive glaciation in Antarctica
started about 35 million years ago. Obviously, it was much warmer globally
and locally before each of these events (until you go back about 200
million more). The wobbles in the Earth's orbit (Milankovitch cycles) do
have some effect on the climate, changing the extent of glaciation but not
changing "generally cooler than the geologic average" or "generally warmer
than the geologic average". Within the past 2 million or so years, the
climate has occasionally been warmer than today, but not by very much.
It should be possible to get a pretty good idea of the climate associated
with the early hominids in question by studying the associated pollen and
other fossils. This would not resolve the question of seasonal migration,
but in some places even the summers may not appeal to poorly insulated
people. Is there any data on how far modern people travel without horses
or wheels, to see if they could reasonably reach warmer climates in winter?
It may also be possible to do direct measurements on the human specimens
of oxygen isotopes. If they stayed put at high latitudes, their bones
would have a fairly constant, low level of 18O (with some seasonal
variation), similar to other bones of non-migratory animals. If they
migrated over long distances, there would be correspondingly greater
seasonal variation in the isotopes, and marked differences with other
organisms. It's been done for elephants, but I do not know of such a study
on hominids-it is a bit destructive.