Machine Learning Algorithms for Transplanting Accelerometer Observations in Future Satellite Gravimetry Missions

2025-08-05

AI Summary

Future satellite gravimetry missions face limitations in electrostatic accelerometer performance, compromising data continuity and measurement accuracy. Method: We propose a novel quantum–data-driven hybrid solution: (i) a dual-sensor configuration integrating a cold-atom interferometer with a conventional electrostatic accelerometer, and (ii) a neural-network-based accelerometer data transplantation algorithm, evaluated in a closed-loop low-low satellite-to-satellite tracking (LL-SST) simulation framework. Contribution/Results: The hybrid configuration significantly improves gravity field recovery accuracy; the data transplantation scheme achieves near-optimal performance with minimal hardware modification, substantially enhancing mission robustness and cost efficiency. This work establishes a scalable paradigm for high-precision monitoring of Earth’s mass redistribution—including glacier melt, hydrological variations, and crustal deformation—and advances the deep integration of quantum sensing and machine learning in space geodesy.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Quantum Machine Learning
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Security and Privacy: Data transparency and provenance
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Accurate and continuous monitoring of Earth's gravity field is essential for tracking mass redistribution processes linked to climate variability, hydrological cycles, and geodynamic phenomena. While the GRACE and GRACE Follow-On (GRACE-FO) missions have set the benchmark for satellite gravimetry using low-low satellite to satellite tracking (LL-SST), the precision of gravity field recovery still strongly depends on the quality of accelerometer (ACC) performance and the continuity of ACC data. Traditional electrostatic accelerometers (EA) face limitations that can hinder mission outcomes, prompting exploration of advanced sensor technologies and data recovery techniques. This study presents a systematic evaluation of accelerometer data transplantation using novel accelerometer configurations, including Cold Atom Interferometry (CAI) accelerometers and hybrid EA-CAI setups, and applying both analytical and machine learning-based methods. Using comprehensive closed-loop LL-SST simulations, we compare four scenarios ranging from the conventional EA-only setup to ideal dual hybrid configurations, with a particular focus on the performance of transplant-based approaches using different neural network approaches. Our results show that the dual hybrid configuration provides the most accurate gravity field retrieval. However, the transplant-based hybrid setup, especially when supported by machine learning, emerges as a robust and cost-effective alternative, achieving comparable performance with minimal extra hardware. These findings highlight the promise of combining quantum sensor technology and data-driven transplantation for future satellite gravimetry missions, paving the way for improved global monitoring of Earth's dynamic gravity field.

Problem

• Improving gravity field recovery accuracy using advanced accelerometer technologies
• Addressing limitations of traditional electrostatic accelerometers in satellite missions
• Evaluating machine learning for accelerometer data transplantation in gravimetry

Innovation

• Using Cold Atom Interferometry accelerometers for precision
• Hybrid EA-CAI setups with machine learning
• Transplant-based approaches for cost-effective solutions