Resident Space Objects Tracking Using Estimation-Based Data Fusion

In recent decades, the population of space debris has increased exponentially, impacting the sustainable development of near-Earth space operations. At present, several commercial space actors are considering deployment of small to medium-sized satellites, with an estimated count exceeding 20,000 by 2035. The prevailing statistics of the space domain pose a significant challenge for satellite operators, underscoring the necessity for sophisticated real-time orbit determination methodologies in the context of Space Domain Awareness (SDA). Typically, ground-based sensors are used for tracking and surveillance of Resident Space Objects (RSO). However, Space-Based Space Surveillance (SBSS) emerges as a promising approach to enhance existing capabilities by offering superior detectability, accuracy and independence from atmospheric conditions. Exploiting multiple optical sensors is imperative for precise RSO tracking, as a single sensor proves insufficient to accomplish the task due to limited Field of View (FOV) and observation time constraints. A Distributed Satellite System (DSS) architecture is discussed for a SBSS mission equipped with Electro-Optical (EO) sensors as piggy-backed payload and inter-satellite communication links to interact, communicate and cooperate with each other to accomplish optimized RSO surveillance tasks. More specifically, this paper proposes an SBSS tracking method that integrates angle data derived from image sequences captured by multiple EO sensors, through an Extended Kalman Filter (EKF). The performance of this method is evaluated across various simulated tracking scenarios. Fusing the data from multiple optical sensors is beneficial as the data from one camera view complements the data from the other camera view to obtain enhanced target measurement information resulting in accurate RSO state estimates. Results show the efficacy of the proposed tracking technique and highlight the opportunities for augmentation of conventional ground-based sensor networks with SBSS paving a path for future research in this field.