Probabilistic Artificial Intelligence Framework for Data Association in Distributed Multi-Sensor Space Object Tracking under Sparse Measurement Regimes

This dataset supports research on a probabilistic artificial intelligence framework for data association in distributed multi-sensor space object tracking under sparse measurement conditions. It contains synthetically generated multi-sensor observations (range, Doppler, and azimuth) produced via Monte Carlo simulations (1000 runs), modeling realistic space surveillance scenarios with noise, clutter, and missed detections. Ground truth trajectories are provided for validation. The dataset enables benchmarking of AI-based data association methods and supports applications in space situational awareness, multi-target tracking, and sensor fusion under uncertainty. Limitations of this dataset include its synthetic nature, which may not fully capture real-world sensor biases, environmental effects, or adversarial conditions. Noise and clutter are modeled using simplified statistical assumptions, and only a limited set of sensor modalities is considered. Additionally, the framework has been evaluated in offline simulations and not in real-time operational settings, and scalability to highly congested orbital environments requires further investigation. All data and scripts necessary to reproduce the results are included in this repository.