...would be for an asteroid (or comet) whose orbit is already fairly well known have its orbit unexpectedly perturbed by a close encounter with the moon or earth or perhaps even some other solar system object. This slightly modified orbit might not be a threat to us immediately, but decades, perhaps centuries from now, on some future flyby, might bring it into close proximity with earth.

The solar system has a long history, and many of the potentially hazardous rocks flying around it have either already collided with some other planet (maybe even us!) or have been ejected out of the system altogether. Others have been kicked into high parabolic orbits that may not bring them back into the inner solar system for centuries. So the system appears to be pretty stable and safe. But every now and then, there is a gravitational encounter that alters an existing orbit, and that alteration may suddenly make it a threat, or direct it into yet another encounter that might make it a threat way down the line.

For example, a comet like Halley's with a well charted orbit, one that has been observed passing into the inner solar system
numerous times in the past, may be approaching Neptune right now, having its orbit slightly altered so that it will eventually collide with the earth. That collision may not be on its next approach to us, but may be in the distant future, perhaps a dozen passes through our neighborhood later.

We would not see this comet as an immediate threat to us, but careful measurements of its orbital elements might alert us to the fact that its orbit has changed, however slightly, and we need to do a lot of number crunching to determine whether or not it will become a threat in the future. Fortunately, if we know well ahead of time a threat is coming, we have the ability to start working on an appropriate response to it. Knowing we are on a collision course with a nearby object is of little use to us unless it is a small object of limited destructive potential. Our only response would be to hurriedly evacuate those populated areas it was likely to affect.

Truly large objects, capable of planet-wide catastrophic damage, would require long periods of time and large technological investments to implement a defensive strategy (like nudging it into a safer orbit or destroying it). We're talking planetary mobilization here, like fighting a World War or stopping climate change.

A truly effective planetary defense system would require a detailed inventory of every object orbiting the sun capable of both damaging the earth and potentially being able to hit it. And that catalog would need to be continuously updated. Although this survey would eliminate lots of objects as potential hazards, it would still leave many more that require further monitoring. To use a maritime example, we're not charting reefs here, we're tracking icebergs. The two tasks are related, but not identical.

The system described in this NASA proposal might be of use in the event a small collider was the threat, and there is nothing wrong with this, but we should also try to design this first step as one that can be economically scaled up to deal with the real planet busters and dinosaur killers that are lurking out there.