SteamKing said:
In the early days of the space program, several orbital missions above the Earth were devised to practice docking the manned capsule with an unmanned craft, both of which were put into the same (or nearly the same orbit). When the astronauts first approached their target docking, they found that there was a difference in vertical separation, w.r.t. the view of the astronauts in the manned capsule. The astronauts tried using attitude thrusters to bring the two craft into alignment. This maneuver failed, and the problem was not resolved until someone on the ground realized that only by changing the speed of the manned capsule (faster or slower, depending on the relative positions) could the vertical separation of the manned and unmanned vehicles be eliminated.
The general problems of orbital rendezvous have been understood for a long time. For objects which are close together, the immediate effect of thrusters is as intuitively expected, but the complications are related to time scales which are a significant fraction of the orbital period, which for low Earth orbit is around 90 minutes. This means that relative velocities can change significantly over a few minutes just as a result of orbital effects.
For example, if you have two objects in a similar circular orbit, and one fires thrusters to move upwards, that will have the effect of making its orbit elliptical, first going higher then going lower than the original orbit. Paradoxically, if you boost forward in orbit, that has the effect of raising the orbit which increases its period, so over a whole orbit you will drop back relative to something ahead of you in the original orbit, and vice versa.
If you look at the relative motion as a pair of objects move around an orbit, a forward boosted object will soon start to change towards upwards motion, then into backwards motion, then downwards in a loop, ending up coming round to forwards again at the original altitude but further back in the orbit. Similar looping motions apply for other small differences of velocity.
In Newtonian mechanics, a particle which passes through a particular point in an orbit around a static spherical object and is in free fall will pass through the same point on its next orbit, if there is a next orbit (that is, unless it escapes completely or hits the atmosphere). The "same point" in this sense ignores the rotation of the Earth, so it will not necessarily occur above the same location on the Earth, and the time taken to get back to that point depends on the energy of the orbit, so particles given different energy by the same explosion will pass through the same point later (if they have not hit the atmosphere) but at different times.