Are robots more cost effective for science work on Mars than astronauts? It may seem a bit OffTopic, but is perhaps a good exercise in project budgeting. Some excerpts from a debate on another discussion group:

''The robots cannot make decisions on the fly, other than extremely simple obstacle avoidance.''

For the same cost as astronauts, we can have 20 or more robots with high bandwidth at 20 different locations. And, they can stay there a long time, unlike astronauts (unless we build a very expensive base). The Tortoise wins this race in the end.

''An astronaut can walk faster than these robots can move.''

20 robots over 4 years are going to do more science than a couple of humans can in a month. And, cover a wider variety of territory.

''A few astronauts and you can do as much exploration in a day as the Spirit and Opportunity have done in their entire existence.''

I don't know about that. Some of those spectrometer readings take several hours to perform even if a human is there. With more money, some of that would happen a lot faster. But power on Mars is going to cost money regardless of whether it is produced for humans or robots. Add to that knowing human life support is a huge energy hog.

Further, the rover operators have been very cautious. If they were less cautious, then more can happen in a day. We just may have to live with losing say 3 out of 20 robots to "go for it".

What would really be helpful is sample returns enabled by robots. The problem is the potential biological contamination. But this issue if faced by both scenarios.

And, Spirit and Opportunity are still mostly low-end robots. With more funding, fancier ones can be built, and still be much cheaper than humans. Here is a summary of ways to beef them up:

* More bandwidth to Earth ''See PlanetaryInternet''
* More power (either bigger panels or "nuke" packs)
* More instruments
* Take more risk
* Improve auto-guidance (more R&D)
* Sample returns
* Multiple "arms"
* Night-time light
* More powerful cameras
* A separate more nimble "survey" bot that finds targets for the lab rover
----
NASA is looking at a variation of the CanadArm 2 (used on the IIS) to service Hubble instead of humans. We can then compare the results with the famous rescue missions/spacewalks of StoreyMusgrave et al.
http://www.space.com/news/hubble_mdrobotic_040615.html to see how they compare. Personally I think we should do both - humans following robots. We've sent several robots to Mars we should now send a few humans even if it is just symbolic, like the Apollo Missions to the moon. But we should look for obvious gaps in technology deployment/monitoring here on earth, such as Tsunami warning systems (and biological, and chemical, asteroid/comet detection etc) so there will be humans left to explore the other planets when the time comes.
----
If we use robots then we can use micro-spaceships. Little paper planes specifically targeted at aspects of exploration. Maybe sent out in groups. Maybe sent out as an interactive network. Massively redundant. (The seagulls had better watch out at the launch site)
----
CategoryDecisionMaking