Neuromorphic Vision-Based Dataset for Space Situational Awareness Applications

Space debris poses growing risks to future space missions and operations that intensified the need for advanced Space Situational Awareness (SSA) technologies. Neuromorphic vision sensors (NVS) offer a potential alternative to conventional optical and radar ground systems, providing microsecond temporal resolution, high dynamic range, and high altitude observations.

In this work, we present a new labeled dataset for real-time detection and tracking of Resident Space Objects (RSOs). The data is acquired using two neuromorphic sensors DAVIS346 and DVXplorer640 integrated with a high-precision 0.8 m aperture Ritchey–Chrétien telescope. The dataset includes event-based recordings of 15 satellites and space debris, along with stars of varying brightness for performance benchmarking.