Particle Filter with Novel Nonlinear Error Model for Miniature Gyroscope-Based Measurement While Drilling Navigation.

Area of Science:

• Geophysics
• Navigation Systems
• Sensor Technology

Background:

• Conventional error models for miniature gyroscope-based measurement while drilling (MGWD) systems rely on small-angle attitude approximations.
• Simplifying the direction cosine matrix (DCM) can lead to navigation errors, particularly with microelectromechanical system (MEMS) sensors in challenging downhole environments.

Purpose of the Study:

• To propose a novel nonlinear error model (NNEM) that incorporates the error of the DCM.
• To reduce propagated errors in MGWD systems, especially under large-angle attitude error conditions.

Main Methods:

• Developed a novel nonlinear error model (NNEM) introducing DCM error.
• Utilized zero velocity and zero position as reference points for state estimation.
• Employed particle filter (PF) and Kalman filter (KF) for state estimation.

Main Results:

• The proposed NNEM effectively reduces propagated errors under large-angle attitude conditions.
• Particle filter (PF) demonstrated superior performance compared to the Kalman filter (KF).
• PF with NNEM significantly restrained system state errors (azimuth, velocity, height) in quasi-stationary conditions.

Conclusions:

• The novel nonlinear error model (NNEM) enhances the accuracy of miniature gyroscope-based measurement while drilling (MGWD) systems.
• Particle filter (PF) combined with NNEM offers a robust solution for navigation in harsh downhole environments.
• The NNEM is particularly effective in mitigating errors associated with large attitude deviations.