Before embarking on settlement and exploitation missions, the distant targets need to be investigated. It would be sad to spend major bucks developing one NEO only later to learn that another one has better concentrations of whatever you are interested in as well as being in a more convenient orbit. It would also be unfortunate to focus on Mars for instance only to learn that another destination suits your purpose better, and vice-versa.
One solution is to send a lot of inexpensive probes to every destination that just might be of interest to you, or anybody that might be willing to buy the information from you. Cassini was impressive, at the Saturn system, told us nothing about about anywhere else. In order to exploit and settle the solar system, we need a wide body of information about everywhere.
Advances in electronics and communications have reached the point that a 10 kilogram vehicle of 2017 is far more capable than one of the massive missions of yesteryear. The trick is keeping them affordable, and transporting them to places of interest. I suggest that the answer to both could be a fairly small rotovator. A rotovator in an eccentric orbit to Lunar distance would be at just under escape velocity at perigee. It could pick up a probe with no significant propulsion from LEO and sling it to well above escape velocity with no onboard propellant used. The rotovator could recover lost momentum with high efficiency electric propulsion. In this way, the probe uses a multi-ton propulsion system that it leaves behind for future use.
A 3 km/sec rotovator is a bridge too far for a first generation system. A solution is to find less challenging work to learn on. A thorough investigation of the Van Allen belts might be an early challenge for the adolescent rotovator/probe system. A group of small probes is carried into LEO as a secondary payload. The rotovator with just a few hundred m/sec intercept velocity picks one up and slings it to an eccentric orbit through the belts. Reboost and repeat as often as feasible.
Next step is to raise the eccentricity of the rotovator a bit more and start sending probes to GEO and Lunar orbit. Repair vehicles and tugs to GEO could be a market as well as units dedicated to nudging dead sats out of GEO into a destructive reentry orbit. The ones to Lunar orbit could get low and expendable to map with a precision only dreamed of today. Other probes could hit the L points for various reasons.
When the rotovator/probe system is more mature the eccentricity is raised again to nearly escape for the distant probes mentioned earlier. One a week departing at almost 3 km/sec above escape could explore most of the solar system with just a little gravity assist. Fifty deep space probes a year should get the job done, or more rotovators could be orbited to up the tempo. One thing that would be explored here is the fast passages to Venus and Mars as this is a much hotter velocity that the propellant limited exploration vehicles to date.
The rotovator itself should mass about a ton. The solar panels and electric propulsion about two more. Total mass being under four tons and quite compact in launch configuration, it might ride share to orbit on an F9, Atlas, or Ariane.
The rotovator experience would be valuable in itself of course. One in Lunar orbit could land loads from Earth and pick up stationary loads from the surface without burning propellant. Others at Mars, Venus, and scattered through the solar system coe the transportation hubs only dreamed of t the present time.