• data fusion
• multi-bistatic radar
• orbit determination
• passive radar
• resident space object
• unscented Kalman filter

The purpose of this thesis, at the end of this research period, is to describe algorithms for tracking and orbit determination of resident space objects (RSO) moving on the Low Earth Orbit (LEO) by using radar and optical sensors. First of all a study of physical model has been carried out, in order to characterize the dynamics of the problem and to identify the forces acting on an object in orbit. After this physical characterization, radar technologies has been analyzed. Particularly, a multi-bistatic configuration has been considered, thanks to the opportunity to use system realized for different purposes, such as radiotelescope, which can be used as radar receivers.
Once analyzed the multi-bistatic radar configuration, different techniques for tracking and initial orbit determination (IOD) methods has been analyzed. The proposed optimal approach consists into a two-step method in which, considering radar range measurements from each receiver, position vector is estimated by using a multilateration algorithm; then, by using an Unscented Kalman Filter (UKF) the complete state vector of the satellite, consisting in position and velocity vector, is estimated. The use of UKF allows also to estimate covariance matrix of the state vector, so that also a statistic information is known.
Once analyzed the feasibility of such system, a wide scenario has been considered, in which the multi-bistatic radar configuration is part of a sensor network, made up of both radar and optical sensors. At this purpose the data fusion problem has been analyzed and UKF has been used in this case both for tracking and data fusion, since it allows the use of measurements of different type to estimate the state vector.
A further analyzed problem, once considered the scenario of a sensor network, has been the optimization of such network and the identification of an optimal scheduling algorithm to minimize the allocation cost. For this reason a new cost function has been identified and validated.
A final topic analyzed during this research activity has been the use of a space-based passive radar configuration in which GEO video broadcasting satellite could be used as illuminator of opportunity. This scenario allows the realization of a low cost sensor network by using Commercial Off-the-Shelf component (COTS) for the design of receivers and already existing transmitters.
As a result of this research activity and as follows from the listed publications, the main author contribution to the state of the art regards research about the feasibility of a multi-bistatic radar system and preliminary results about the feasibility of a passive radar system for Space Surveillance and Tracking (SST).
In conclusion, the innovative contribution of this thesis will be:
- Multi-bistatic radar system: multi-bistatic radar configuration has been considered in order to perform initial orbit determination of Resident Space Objects. For this purpose radiotelescopes has been considered as radar receivers, so that already existing assets can be used.
- Radar and optical measurements data fusion: in order to improve the estimation performances both radar and optical sensors has been considered and a data fusion algorithm has been developed.
- Multisensor scheduling optimization: In a realistic scenario more than one sensor is available in order to perform tracking and detection of RSOs, for this reason an optimal scheduling of this sensor network is required.
- Space-based passive bistatic radar (PBR): Geostationary videobroadcasting satellite has been taken into account as illuminator of opportunity for a passive bistatic radar scenario. The known and stationary position of such satellites combined to wide band signals could guarantee Low Earth Orbit RSOs observability.