In-transit space object recognition via physics-based remote sensing

Identifying, classifying, and monitoring space launch vehicles in transit is central to Space Domain Awareness (SDA), informing defensive decision-making to critical space-based assets in congested and contested orbital regimes. This work introduces a physics-driven methodology for generating multi-modal synthetic datasets, integrating transportation vehicle dynamics, material properties, environmental factors, and correlated EO/IR, LiDAR, and SAR signatures. The resulting observables capture formations, spatial extent, and geolocation appearances across diverse sensor perspectives, providing the variability needed to train context-aware AI systems. By spanning diverse operational scenarios and logistical configurations, the dataset enables machine learning models to generalize beyond empirical baselines and operate effectively under dynamic conditions beyond peacetime settings. These capabilities extend from ground-based telescopes to aerial platforms, enabling predictive assessment and indicators of object behavior, system readiness, and operational performance. For defense applications, the approach strengthens mission planning, sensor tasking, and object custody, while also revealing observational gaps and informing architecture design. Ultimately, physics-based modeling enhances the fidelity of threat assessments and fortifies resilience against adversarial actions, empowering operators to sustain persistent awareness and protect critical assets in modern space operations.