This model determines whether or not an object in space has propulsion. The model uses a Two Line Element (TLE) set as input. As output, it provides a classification of whether or not an object in space has propulsion. This model can be used to determine whether an object in space is debris or a potential adversarial satellite; or used as an input to assess whether a satellite may potentially collide with one or more other objects in space, or determine whether a satellite can be moved to observe specific regions of the Earth.
Given a TLE set, we extract the Keplerian Elements, inclination, right ascension of the ascending node, eccentricity, argument of perigee, mean anomaly, and mean motion, which are used as inputs to the model. This model uses a Random Forest Classifier to split the data into objects with and without propulsion. The model was trained with labels for each instance for propulsion and non-propulsion objects.
This model was built using a dataset of 3.8M TLEs; after reduction to include only low-earth orbit satellites, the training dataset included 800K records for each of the two classes (propulsion and non-propulsion). Precision and recall metrics were calculated based upon those datasets. This model achieved a precision of 99.87% and a recall of 99.91%
A higher precision score indicates that the majority of labels predicted by the model for different classes are accurate. Further information here.
A higher recall score indicates that the model finds and predicts correct labels for the majority of the classes it is supposed to find. Further information here.
This model was trained and validated on a large sample of satellite data, consisting of two-line element set (TLE) records. These were limited to Low Earth Orbit (LEO) satellites. Given a two-line element set, we are able to extract Keplerian elements, inclination, right ascension of the ascending node, eccentricity, argument of perigee, mean anomaly, and mean motion. With these given inputs, the model indicates whether an object has propulsion or not. This model uses a Random Forest Classifier to split the data into object with and without propulsion. The model was trained with labels for each instance for propulsion and non-propulsion objects.
The model was trained using an 85% sample of the 1.6 million TLE records that were narrowed from a larger set of over 2 million to include only LEO satellite orbits.
The performance of the model was tested on a validation dataset consisting of 240K TLE records, a 15% sample from the original set.
Necessary cookies are absolutely essential for the website to function properly. These cookies ensure basic functionalities and security features of the website, anonymously.
Functional cookies help to perform certain functionalities like sharing the content of the website on social media platforms, collect feedbacks, and other third-party features.
Performance cookies are used to understand and analyze the key performance indexes of the website which helps in delivering a better user experience and Modzy product offering.
Analytical cookies are used to understand how visitors interact with the website. These cookies help provide information on metrics the number of visitors, bounce rate, traffic source, etc.