The present work focuses on the investigation of an artificial intelligence (AI) algorithm for brightest objects segmentation in night sky images’ field of view (FOV). This task is mandatory for many applications that want to focus on the brightest objects in an optical sensor image with a particular shape: point-like or streak. The algorithm is developed as a dedicated application for star sensors both for attitude determination (AD) and onboard space surveillance and tracking (SST) tasks. Indeed, in the former, the brightest objects of most concern are stars, while in the latter they are resident space objects (RSOs). Focusing attention on these shapes, an AI-based segmentation approach can be investigated. This will be carried out by designing, developing and testing a convolutional neural network (CNN)-based algorithm. In particular, a U-Net will be used to tackle this problem. A dataset for the design process of the algorithm, network training and tests is created using both real and simulated images. In the end, comparison with traditional segmentation algorithms will be performed, and results will be presented and discussed together with the proposal of an electro-optical payload for a small satellite for an in-orbit validation (IOV) mission.

The increased risk of collisions between objects orbiting the Earth is a topic of great relevance in the framework of Space Traffic Management (STM). Resident Space Objects (RSOs) are very difficult to detect and identify in terms of orbital parameters due to their variable sizes and the too-short duration of available observations in most cases (Too-Short Arcs problem). An initial orbit determination algorithm which employs a combination of Metaheuristic and Physics-Informed frameworks as backbone and integrates an adaptive filtering module to improve the robustness to random parameter initialization and outliers, has been developed and described in this work The algorithm has been tested on a created dataset of 1800 simulated mission scenarios with random targets and space observers, considering variable observation duration from 20 to 60 s. Results show that in the 99.61% of the scenarios, the reference orbit is retrieved with an accuracy on the final target's state greater than 1 km and 0.001 km/s and a mean RSSE across the observation arc of 0.002 km.