Dual-Layer Factor-Graph Optimization for Delayed Star-Tracker/IMU Fusion in Highly Dynamic Spacecraft Attitude...

Area of Science:

• Aerospace Engineering
• Robotics
• Control Systems

Background:

• Accurate spacecraft attitude estimation is critical for dynamic missions.
• Challenges include asynchronous sensing, motion blur, and delayed data from star trackers and inertial measurement units (IMUs).
• Existing fusion methods struggle with highly dynamic conditions and sensor data inconsistencies.

Purpose of the Study:

• To develop a robust attitude estimation framework for highly dynamic spacecraft.
• To address challenges posed by asynchronous sensing, motion blur, and delayed outputs.
• To improve temporal consistency and accuracy in attitude estimation.

Main Methods:

• A dual-layer factor graph optimization framework is proposed.
• Lower layer: Fuses high-rate IMU data with motion-blurred star streak observations.
• Upper layer: Integrates delayed attitude constraints, propagated star vectors, and inertial constraints.

Main Results:

• The framework demonstrates higher estimation accuracy and robustness compared to existing methods.
• It shows improved tolerance to delayed or intermittent star-tracker observations.
• Computational efficiency is maintained for near-real-time onboard implementation.

Conclusions:

• The proposed dual-layer framework effectively handles asynchronous and blurred sensor data for spacecraft attitude estimation.
• It offers a significant improvement in accuracy, robustness, and temporal consistency.
• The method is suitable for near-real-time onboard applications in highly dynamic scenarios.