Machine Learning-Based Evaluation of Attitude Sensor Characteristics Using Microsatellite Flight Data

TL;DR

This paper presents a machine learning approach using Conv1D neural networks to improve attitude sensor data accuracy from microsatellite flight data, significantly reducing errors compared to traditional methods.

Contribution

It introduces a novel machine learning method for attitude error correction that outperforms conventional physical-model-based approaches using real satellite data.

Findings

Achieved RMS attitude errors of ~0.7 degrees on training data.

Reduced errors to 2-3 degrees on test data.

Demonstrated effectiveness of Conv1D in satellite attitude correction.

Abstract

Using actual flight data from a 50-cm-class microsatellite whose mission and operations have already been completed, this study re-evaluates satellite attitude determination performance and the error characteristics of onboard attitude sensors. While conventional approaches rely on batch estimation or Kalman filtering based on predefined physical models and white-noise assumptions, this research introduces a machine-learning-based approach to extract and correct structural and nonlinear error patterns embedded in real observational data. In this study, high-quality attitude determination results obtained from star sensors and a fiber optical gyro (FOG) are treated as ground truth, and machine learning is applied to coarse attitude sensor data consisting of Sun sensors and magnetic field sensors. A one-dimensional convolutional neural network (Conv1D) is employed to regressively predict attitude from short sequences of time-series sensor measurements. The model is trained and evaluated using five sets of on-orbit observation logs, with four passes used for training and one independent pass used for testing. The results show that, while conventional coarse attitude determination using the TRIAD method yields attitude errors on the order of 7 deg RMS, the proposed machine-learning approach achieves RMS errors of approximately 0.7 deg for the training data and 2-3 deg for the test data, depending on the sensor combination.