Imagine a large body (for example, a planet) in 3D hyperbolic space. Now imagine the planet starts moving in a straight line at constant speed.
In Euclidean space, all points would move along parallel lines. In hyperbolic space, however, most points will be moving along equidistant curves, or hypercycles.
But points moving along a curve should experience some sort of force. If an observer is on the surface of the planet, he should be able to measure larger or smaller force depending on how close he is to the axis of movement.
This leads to conclusion that, unlike Euclidean space, hyperbolic space would allow an observer to determine whether the planet is in rest or whether it is moving without referring to any other body -- absolute rest/movement, in effect.
Is this line of reasoning correct?
Contrary to my initial gut feeling, this sounds highly plausible. If you assume that any body which is not moving (with constant speed) along a geodesic experiences some force, then you can simply take four test bodies, forming a regular hyperbolic tetrahedron, and consider translating that through space. As you said, a hyperbolic translation only moves points on a specific geodesic along that same geodesic, while all other points are moved along curves equidistant to that geodesic. So at most two corners of the tetrahedron can remain without force, at least two others will take a curved trajectory and hence be subject to some force. So I'd say you are right.
Whether the force is sufficiently large to be measured in practice depends of course on the scale of things, so we still don't know whether the universe is flat or not.