In this work we describe JAXA LEO Survey System (JALESS), a space debris monitoring system capable of detecting and tracking uncatalogued space debris at LEO, consisting in a multi-site optical telescope arrangement. The detection process consists in a data-intensive computing technique, aided with hardware acceleration through parallel multi GPU-FPGA-CPU heterogeneous computing, which makes the process enough sensitive to detect uncatalogued space debris (smaller than 10 cm), and enough fast to create in time re-observation coordinates for the tracking sites, based on observations on the detection sites. Our current multi-site infrastructure consists of a detection site at Siding Spring Observatory (SSO) in Eastern Australia, and a tracking site at Zadko Observatory in Western Australia. The system can detect uncatalogued space debris at SSO and re-detect the same object in its next pass over Zadko Observatory. When the system detects an uncatalogued space debris object at SSO, it computes a first orbit approximation, and sends re-observation coordinates in time to the tracking site at Zadko Observatory. Using the observation coordinates of the same object from both sites, the system calculates more accurate orbital parameters, and uses them to track this object during consecutive days. We are currently building up a database of uncatalogued space debris, that we believe can positively contribute to enhance space safety at LEO, as this system can compensate the inability of current space debris monitoring systems to detect and track small uncatalogued space debris.
POLO et al. (2025) studied this question.