Figure 1. The Speed-of-Light Communication Delay Makes Artificial Intelligence a Necessity for Space Exploration.
ration need to be extremely robust to a wide array of
possible disturbances and capable of a wide array of
tasks. In short, as the human race expands its efforts
to explore the solar system, artificial intelligence will
play a key role in many high-level control decisions.
However, giving a rover that cost many person-years of labor and a multimillion dollar budget complete autonomy over its actions on another planet
might be a bit unnerving. Space is a harsh and dangerous place; what if it isn’t able to achieve the tasks
it needs to? Worse, what if the rover finds an unpredicted and creative way to fail? These are legitimate
concerns, worth addressing seriously.
One way to mitigate these concerns is to take the
concept of a single traditional monolithic rover and
broke it up into many pieces, creating a team of
rovers, with one to embody each of these pieces. Each
would be simple and perform just a few functions.
Though each of the pieces is less effective individually than the monolithic rover, the sum of the pieces is
greater than the whole in many ways.
First, any of the members of the team is signifi-
cantly more expendable than the whole monolithic
rover. This alleviates a large number of concerns and
opens many opportunities. If one rover does find a
way to fail creatively, the remainder of the team is
still completely operational. By the same token, the
team of rovers can undertake more dangerous mis-
sions than the monolithic rover; if the dangerous
conditions lead to the failure of one rover, the rest
can complete the mission. Additionally, redundancy
can be designed into the team for particularly dan-
gerous or critical roles.
Beyond the disposability of the individual team
members, there are other benefits to this team-based
approach. Savings can be realized in construction, as
each rover can be designed with parts from a lower-cost parts portion of the reliability curve. Similar savings are available in the design process, as a new team
can be formed with some members that have been
previously designed.
In addition, a team of rovers can have capabilities
that a single monolithic rover cannot, like having
presence in multiple locations at once, which is
incredibly useful for planetary exploration. Ephemeral events can be simultaneously observed from separate locations (Estlin et al. 2010), even from the
ground and from orbit simultaneously (Chien et al.
2011), which can make interpreting the situation significantly easier. Construction tasks that might be
impossible for a single rover with limited degrees of
freedom become much easier. Teams can survey areas
separated by impassible terrain and share long-range
communication resources (Chien et al. 2000).
However, the concerns that we must address
expand rapidly once we start to consider the possibilities that arise with multiple rovers acting in the
same area simultaneously. How do the rovers coordinate so that their efforts lead to the maximum
amount of interesting discoveries? How does a rover
decide between achieving a task on its own versus
helping another rover that has become stuck? How
