• mathematical physics
• NEOs
• dynamics
• impact monitoring

Our solar system is full of different bodies, including asteroids, that represent the majority of
objects we can find and at the same time the greatest risk for our security. We know that since
the formation of our Earth, it has been impacted many times by natural objects, nevertheless
the Impact Monitoring (IM) of Near Eart Objects (NEOs), i.e the objects that are closer to Earth
(so the most dangerous), is a surprisingly young field of research, that started a little more than
20 years ago, when in 1999 the first IM system became operational.
In these 23 years the advance of technologies allowed us to get increasingly accurate results
in the prediction of possible impacts in the future and in the assessment of the risk.
Starting from the observations of an object we are nowadays able to understand the dynamical evolution of the body in the future times and predict accurately enough the possible future
impacts of the object with the Earth, assessing the probability of the impact and even the exact
date of the event, so that we can be prepared to eventually evacuate the population if needed.
What an IM system does is searching inside the so called Confidence Region (CR), a region
of uncertainty in which the astrometric residuals of a preliminary orbit are acceptable, some
Virtual Asteroids (VAs), i.e. the orbits of the possible dynamic evolution of the body we are
studying. The goal of impact monitoring is to understand if the CR contains some Virtual Im-
pactors (VIs), that are subsets of initial conditions that propagated lead to an impact with the
Earth.Then the IM system, once a VI is found, needs to compute the Impact Probability (IP) of
the event, to assess the risk of an actual impact.
The IM systems currently operational are CLOMON2, AstOD, Sentry and Sentry-II: the first
three work using the Line of Variations (LOV), while the last one uses a completely new
method.
In this thesis we will talk about the latest development in IM science, giving a review of the
new technique implemented in Sentry-II, operational at the Jet Propulsion Laboratory (JPL)
of NASA. This new system allows us to get better results in the Impact Monitoring field, over-
coming the results of the previous systems based on LOV, already very accurate, solving some
limitations that the LOV based methods had.
This new method, how we will see, incorporates the impact coordinates in the observation vector, allowing us to march naturally trough the uncertainty region toward impactors, instead
of moving inside the LOV, so that the algorithm results in being more accurate in finding VIs.
The goal of the thesis will be at the end to see, working directly on the observational data
using the software OrbFit, if there is and how much it is the difference between the results
obtained with the ”old” IM systems, whose method was based on the LOV, and the new results
of Sentry-II.
The thesis work is divided as follows:
• Chapter 1: gives us a brief introduction on the tools of Orbit Determination (OD). After
some basics and definitions, the least squares method will be introduced and we will see
what is meant for Identification problem and for Admissible Region (AR), to give at the
end some basics informations on NEOs population.
• Chapter 2: starting from a little recap of the history of the IM, we will see the Classical
Impact Monitoring, i.e. the theoretical formalization of the tools used in the search of
VIs and in the estimation of the Impact Probability (IP) for a certain impactor, concluding
with the criteria for the communication of the risk to the public.
• Chapter 3: this is the main body of the thesis, we will talk in depth about the new
method implemented in Sentry-II, talking about the steps that Sentry-II does to give us
the best estimation of the impactors that can hit the Earth in the near future. We will also
see the results in terms of completeness and computational performance accomplished
by the system.
• Chapter 4: using the OrbFit software, in particular the fitobs.x executable program we
will compare the results of Sentry-II, available on the CNEOS (Center for Near Earth
2
Object Studies) page, to the results obtained using the LOV method, to see operationally
the differences.
• Conclusions: we will summarize the results obtained and we will see the future possibilities in the application of Sentry-II.