- establishing an initial infrastructure on the Moon -


In space advocacy circles there is a recurring argument about who is better for space -- robots or humans.

For the advocates for robots, they argue that robots are cheaper and don't require life support. For the advocates of humans, they argue that humans are much faster than robots and have better judgement.

But these arguments miss the point. Robots are better for some things and humans are better for other things. This page will detail what situations are best for each.

But first, let's clarify the three settings in question. Those are:

  • Exploration
  • Development
  • Settlement

The debate and confusion most often arises from not distinguishing between those settings.

The greater speed of humans notwithstanding, robots are almost always better than humans for scientific exploration. For probes exploring the outer solar system this is a given. Only robotic probes can currently explore places beyond Mars. The radiation during transit and long-duration life support requirements make exploration by humans impossible in these locations.

For the Moon and Mars, robots still easily beat humans for several reasons. On a per-mission basis, they are far cheaper than sending astronauts. If we were to mass produce Spirit and Opportunity rovers for Mars, each mission would cost about $400 million whereas the first human mission to Mars would typically cost about $150 billion (that's billion with a "b"!). So, we could conduct about 375 robotic missions for each crewed mission. Most importantly, those robots could go to 375 different locations whereas humans could only go to one location or just a handful if they were to drive around. Science return is highly dependent upon where one goes. So, the more places one goes the more unique science is returned. And scientific rovers are lasting for many years. So, even though they are slower each telerobot could cover a substantially equivalent distance compared with human explorers.

The argument is made that robots are not smart enough to understand the context of what they are looking at. Until we have true AI, only geologists in the field can interpret what they are looking at. This completely ignores how teams of our best geologists on Earth can be virtually "present" on the Moon or Mars via telepresence. Those robots that we have on Mars are as smart as the scientists who operate them.

However, even though there are strong practical reasons for why robots beat humans for exploration in almost every context, there is none-the-less a strong reason why we shouldn't "let the robots have all of the fun" but should hold them back in favor of astronauts conducting scientific exploration.

For both the Moon and Mars, there could be a very extensive International Exploration Phase provides that the per-seat price was low enough. This extensive phase, driven by the motivation of national pride, could be just the thing needed to increase flight frequency and so further reduce the cost of passenger transport thereby opening the way for private settlement.

It is the position of the Space Development Network advocates that NASA's Commercial Landed Payload Service (CLPS) should limit small "commercial" landers to prospecting and perhaps small-scale demonstrations and NOT to visit exciting science websites prior to international astronauts visiting them.

Space development means building up infrastructure including the systems to utilize local resources and the development of habitats as well.

On the Moon, telerobotics would involve a 2.6 second round trip time delay. This is small enough such that there could be a brief robot-only phase where teleoperations robots could set up power systems, harvest and process ice into set up inflatable habitats and dock equipment into ports, and perhaps even start growing food in the greenhouse. With the time delay actions would be slowed but, operating 24/7, considerable work could be accomplished telerobotically prior to crew arrival.

Prior to crew arrival, dexterous telerobots such as the Robonaut 2 could remove worn parts designed with quick release mechanisms and replace them with new spares. In this way, small mass spare parts could keep telerobots working thereby producing product (e.g. water or metals) at quantities perhaps 100X the mass of the spare part.

After crew arrival, the relationship between lunar crew and robots becomes very interesting. The crew would maintain the telerobots, assemble more of them, and even extract metals from the lunar dirt and process them into more telerobotic parts. The telerobots would be controlled by operators on Earth 24/7. In this way, a few crew could maintain a very large telerobotic workforce and development would scale very large very quickly. Telerobots harvesting a processing metals could expand the number of their fellow telerobots and hence get onto an exponential curve.

Robots cannot do human settlement but they can be an integral part of keeping the settlement running. There would be a division of duties. Everything that the robots could do the robots would do. Those tasks that humans could do much better than telerobots from a dexterity (not logic) standpoint the humans would do.

Telerobots from lunar distance means that they could perform tasks with full human intelligence controlling them. But of a task could be safely, fully automated and done with about the same efficiency of teleoperations then those robots would be allowed to operate autonomously. A hybrid would be supervised autonomy where the teleoperations (perhaps controlling multiple telerobots) gives high-level instructions and the robot determines the individual steps to accomplish those directions. Driving while avoiding hazards is one task that now could probably be fully automated. And if a robot gets stuck, a fellow robot could free it.

Here is a list of the robots that are considered to be needed with a growing, permanent, lunar base:


  • Prospector


  • Ice Harvester
  • Dexterous Telerobot (Repair)


  • Robotic Dog
  • Humanoid (bipedal)
  • CameraBot


  • Road Maker
  • Regolith Excavator
  • Autonomous Hauler
  • Autonomous RV


  • Construction

The Plan for Sustainable Space Development envisions telerobots preparing and compacting a network of Level 1 dirt roads on the Moon. Automated vehicles could drive on those roads and record the ups and downs and any loss of traction while going around corners. This recording could be used by later vehicles so as to safely traverse the road at a maximum but safe speed.

These vehicles could take international astronauts and tourists to regional destinations without having to consume a tremendous amount energy electrolyzing water into propellant for a suborbital hop. Traveling at about 25 mph from a polar base to the equator would take about 2.5 days. There would need to be solar-powered charging stations. Perhaps cargo and crew modules could be transferred between electric undercarriages so as to eliminate wait times.

But also, these vehicles could be used to transport water and other volatiles from the poles to bases or settlements in other parts of the Moon. Likewise, autonomous vehicles could transport mined materials such as potassium and phosphorus fertilizer from mines to other bases including to the poles. In this way, no parts of the Moon would be without the resources they need for settlement.

With a well thought-out plan, robots could play a key role in space exploration, development, and settlement.

Next: Polar Resources