This is beyond my knowledge of science and engineering, however this post did get me to consider things I haven't in the past. For example, I had always considered space expolration something that's done with a very large object. Never had it occurred to me that something the size of the coffee cup could be propelled in to space in order to retrieve data.
There's quite a bit of research going into miniature thrusters, which will be very useful for heavily miniaturized probes. Of course, nothing stops us from using these thrusters on larger probes (if you need bigger payloads), and they'd likely be more efficient in that case. It's all about mass ratios and specific impulse, also known as the tyranny of the rocket equation. 100 m/s from a 2x10x10 cm system is quite respectable. Let's take a look at the claim about being able to reach lunar orbit with 6U worth of these (30 systems). You need around 4000 m/s to get from LEO to LLO, or at least 40 of these systems. But this is with having multiple sets of thrusters and not simply having multiple sets of fuel tanks that you jettison along the way. So I recon it's possible, but would likely be tricky due to the low thrust. Perhaps some kind of low-energy transfer could help, like what was done with PAS-22.
We are moving towards larger arrays of smaller spacecraft measuring the same things simultaneously to build a more global picture of in-situ derived space science.
What a great idea. First thing that pops into my mind would be trying to measure something like gravity waves. Do you have any other examples?
Yep, and that idea will eventually see construction, but it's been shelved because it costs money. The four MMS Spacecraft fly in a tetrahedral formation to gather 3D data. With one spacecraft, you have a point; two, a line; three, a plane/2D; four, 3D. STEREO is kinda cool because they use the Earth as another observation vantage of data, in tandem with other input from ACE. In fact, one of the coolest semi-recent developments is people working on solar wind mapping using data from as many current missions (in the solar wind environment) as possible. Even little sounding rocket projects use ground-based observations to cross correlate on-board instrument measurements with, but I digress. The moral of the story is that you just can't have too many datapoints. So now, everyone in the industry is currently developing methods of miniaturizing existing and emerging technologies to save mass, power, payload space, and $'s. Combine that with the cubesat/microsat/nanosat developments, and it's just a logical progression; swarms of several hundred spacecrafts that are equipped with better resolution everything instruments. But hold on there, Hubski isn't /r/futurology. No, our optimism is tempered with at least some measure of realism. NASA will continue to preside over our progression, so you can expect things to continue at a crawling pace. Some issues are solvable with more money thrown at them, but not all of them. Why? Something something bureaucracy, to sum it up. I'll cut myself off here. Also, welcome to Hubski. :)...trying to measure something like gravity waves.