The feasibility of improving the Earth collision uncertainty knowledge of a potentially Earth impacting asteroid by means of a multi-flyby mission has been researched in this thesis. The feasibility is assessed both from an astrodynamics point of view and from a systems engineering point of view. The astrodynamics part provides knowledge evolution results from a covariance analysis, which combines radiometric measurements of the spacecraft from Earth and optical measurements of the asteroid from the spacecraft. The systems engineering aspect on the other hand identifies all mission and subsystem requirements and provides possible subsystem design options.

From the 1906 potentially hazardous asteroids known at the time of writing, Apophis has been selected as the main research target due to its extremely close expected Earth encounter on April 13th, 2029: the expected miss distance is just 38440 km. Asteroids 2012 UE34 and 2001 WN5 have been selected as secondary targets. High-fidelity resonant trajectories have been generated using a multiple shooting method implemented in NASA's General Mission Analysis Tool. As a robust initial guess for these simulations, results from porkchop plots have been used, which combine the solution of Lambert's problem and resonance data. Only 1:1 resonance has been researched for this thesis. All trajectories target the asteroid B-plane at 600 km distance.

Data from JPL's HORIZONS has been used, alongside the generated high-fidelity trajectories, as input for a covariance analysis to produce the asteroid position knowledge evolution and the knowledge evolution of the asteroid in Earth's B-plane at the expected close-approach date. It has been found that, for Apophis, two flybys are sufficient to ensure mission success. After the second set of optical measurements, the orbital period of the asteroid is accurately known, which improves the asteroid position knowledge by roughly two orders of magnitude while improving the diameter of the covariance error ellipse at the expected close-approach date by roughly three orders of magnitude. The 1-sigma error ellipse diameter shrinks from roughly 65000 km to roughly 15 km after two flybys. The analysis shows that for the used spacecraft trajectory the asteroid knowledge evolution is independent from the targeted flyby B-plane parameters. In the case multiple optical measurements are taken during a flyby, the knowledge evolution is found to be close-to independent from the distance to the asteroid when an optical measurement is taken. Asteroid WN5, which does not have a near-circular orbit around 1 AU such as Apophis and UE34, has a different knowledge evolution that does not improve the orbital knowledge after two flybys, meaning that the proposed mission would not be justified for this target.

The major conclusions that result from the systems engineering analysis are the following: a feasible launch strategy would be to use a light launcher, such as Vega, to launch the spacecraft to LEO, in combination with an upper stage, such as STAR 48, to escape Earth's gravity and put the spacecraft in the desired interplanetary trajectory. Feasible options for the propulsion system are an ion thruster, a cold gas system and a hydrazine system. All three require a low velocity increment, in the order of magnitude of a few m/s, which translates into a propellant mass of maximum 0.2 kg. The attitude control system needs to provide a slew rate of at least 0.835 degrees per second during the flyby phase for accurate asteroid pointing in order to make optical measurements. The estimated necessary average power equals 215.16 W and the peak power 485.76 W. The average power will be provided by GaAs solar panels with an estimated area of 0.6709 m^2, while Li-ion batteries with a storage capacity of 3301.11 Wh will be used to provide the necessary power during peak power scenarios. The high-level spacecraft design has an estimated dry mass of 149.77 kg, excluding the mass of the upper stage.

The combination of the astrodynamics aspects and the systems engineering aspects indicate that a flyby mission to a potentially Earth impacting asteroid is a promising option to improve the Earth collision uncertainty knowledge of that asteroid.