There is a common (if not popular) belief that we don’t need humans to explore the solar system when we already have robots to do the work at a lower cost. This opinion is misguided and outright wrong according to the data.
It’s misguided because it leaves out the actual ratio of scientific productivity to cost, which humans are far better than robots on. It’s wrong because it ignores the actual rising cost of robotic planetary exploration, which has occurred despite copious advancements in computer processing power.
Humans vs. Robots in General
Before we can quantitatively assess the benefits of humans vs. robotics for space exploration, it helps to qualitatively understand what each brings to the table. The pros of humans can be summarized as follows:
- On the spot decision making and flexibility, with increased opportunities for making serendipitious discoveries
- Greatly enhanced mobility and attendant opportunities for geological exploration and instrument deployment (compare the 35.7 km traversed in three days by the Apollo 17 astronauts in December 1972 with the almost identical distance (34.4 km) traversed by the Opportunity rover in eight years from January 2004 to December 2011)
- Greatly increased efficiency in sample collection and sample return capacity (compare the 382 kg of samples returned by Apollo with the 0.32 kg returned by the Soviet robotic sample return missions Lunas 16, 20, and 24, and the zero kg returned to-date by any robotic mission to Mars)
- Increased potential for large-scale exploratory activities (drilling) and the deployment and maintenance of complex equipment
- The development of a space-based infrastructure capable of supporting space-based astronomy and other scientific application (the construction and maintenance of large space telescopes)
Their cons can be summarized up in one bullet point – they need lots of attendant mass for life support, pressure vessels, additional equipment, and spare parts. The additional mass translates to vastly increased launch costs. Obviously, robots have exactly the opposite problem. One could take away from this the sense that humans are efficient but expensive, while robots are cheap but inefficient.
So the question becomes this: how much more cost-efficient are humans over robots, even when taking in consideration technological advances made since the Apollo days?
For scientific productivity, look no further than the six successful Apollo landings on the Moon, compared to the Soviet Luna sample return vehicles and Lunakhod rovers. As previously mentioned, all six Apollo missions returned a total of 382 kg of lunar rocks and soil to Earth. Most of them are in a secure, climate controlled bunker at Johnson Space Center, occasionally loaned out to universities and other organizations for their research.
You’d think that, after more than 40 years, the publications from the Apollo missions would have levelled off. But as Figure 1 shows, the number is vast and still rising while the cumulative publications from the lunar robotic missions are small and levelling off. The interesting part is that this was the result of only 12.5 days on the lunar surface, compared to 436 active days by the Lunakhod rovers and 5162 days on the Martian surface by the Spirit and Opportunity rovers. The Apollo astronauts collected samples from over 2000 discrete sites in only 3.4 days of total EVA time, a feat which our most advanced rover, Curiosity, remains incapable of.
Superficially, the argument for robotic cost-effectiveness appears solid. The entire Apollo program cost about $25 billion in 1966 (when Apollo expenditure peaked), which would have been $175 billion today. The Curiosity rover cost $2.5 billion in today’s money. So in real terms, Apollo cost 70 times as much as Curiosity. But Apollo visited six sites to Curiosity‘s one, so in terms of cost per site Apollo was only 12 times as expensive.
That does not take in account the heavy, complex equipment that can be deployed during a human mission, such as deep drills. In light of Apollo’s vast scientific return – two or three orders of magnitude over the lunar robotic missions – the additional cost appears miniscule.
It’s worth noting that the cost of planetary robotic exploration has actually increased over time, despite small advances in autonomous control that are hampered by the need to radiation-proof spacecraft computers. The Pathfinder mission in 1997 had a small rover the size of a shoebox called Sojourner. It was more of a test than an actual scientific tool, and operated for 90 sols on Mars, three times longer than its design life. The total mission (both lander and rover) cost $280 million in 1997 dollars, or $423 million in 2016. Curiosity, a much larger and more capable rover, cost $2.5 billion and is still running after four years. The data would seem to show that larger rovers are preferred to small rovers for their payload capability and range. No matter how much artificial intelligence we pack into a small rover, it does not matter if that rover is too small to do much.
Synergy: An Optimal Exploration Strategy
This is not to show that robots are useless. On the contrary, they (especially when upgraded with more than rudimentary artificial intelligence) can serve as precursors and expendable assets that increase the overall scientific return of a human mission. They can be sent into areas that are too dangerous for human astronauts to venture into, or used for purposes that are too boring/repetitive for humans. Examples might include cliff-traversing rovers, cave-bots, and automated weather/seismic stations.