Research on orbit prediction based on multi-source satellite orbit characteristics and multi-head attention mechanism-LSTM model
Year: 2026
Type: article
Abstract: ABSTRACT High-precision orbit prediction is critical for stable navigation system operation and space safety. To address the limitation of existing data-driven approaches that rely on single-satellite historical data and fail to fully exploit shared information among similar satellites, this paper proposes an orbit error correction model integrating multi-source orbital features with a multi-head attention mechanism-enhanced LSTM. Random forest regression and permutation importance analysis are employed to select SGP4 position errors, Q4 and Beta Angle as core input features, thereby mitigating redundant-feature interference. Unlike single-satellite modeling, the proposed method trains the model using historical data from multiple BeiDou satellites with similar orbital characteristics and applies it to error prediction and orbit correction for BeiDou-3M14. This strategy enables the model to capture satellite-specific error evolution and common orbital error patterns across similar satellites. Experimental results show that residual rates in the X, Y and Z directions are 0.24%, 0.16% and 0.20%, respectively, with root mean square errors of 1.17 m, 0.96 m and 0.75 m, outperforming Long Short-Term Memory(LSTM), Back Propagation Neural Network(BP) and Support Vector Machine(SVM) methods. Further experiments analyze the effects of training-satellite number, neural units and sampling interval on model performance, and verify its applicability to different prediction durations, space targets and GLONASS. The results demonstrate that multi-head attention mechanism-LSTM(MHALSTM) improves SGP4 orbit prediction error correction and supports high-precision space target prediction and space situational awareness. However, its performance remains constrained by orbital similarity, sample size and complex low-Earth-orbit perturbations, and its long-term stability in complex environments requires further verification.
Source: Array
Language: English
Subfield: Aerospace Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,680
Awards: G3288883886, G4613958513
OpenAlex ID: W7214714229
