Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Towards Learning-Based Gyrocompassing

Daniel Engelsman, Itzik Klein

TL;DR

This work tackles the challenge of gyrocompassing with low-cost MEMS gyroscopes by leveraging a learning-based approach to compensate inherent sensor errors during stationary self-alignment. A Bi-LSTM framework is proposed, paired with a cyclic mean squared error loss (CMSE) and data augmentation, to estimate the heading angle from gyro streams efficiently. Experimental results show the method can reduce the required alignment time by up to a factor of about 10 and lower heading error by more than 50% on unseen data, bringing practical gyrocompassing within reach for affordable sensors. The findings suggest that software-enabled gyrocompassing can extend high-end navigation capabilities to off-the-shelf instruments, with wide relevance for autonomous and tactical platforms.

Abstract

Inertial navigation systems (INS) are widely used in both manned and autonomous platforms. One of the most critical tasks prior to their operation is to accurately determine their initial alignment while stationary, as it forms the cornerstone for the entire INS operational trajectory. While low-performance accelerometers can easily determine roll and pitch angles (leveling), establishing the heading angle (gyrocompassing) with low-performance gyros proves to be a challenging task without additional sensors. This arises from the limited signal strength of Earth's rotation rate, often overridden by gyro noise itself. To circumvent this deficiency, in this study we present a practical deep learning framework to effectively compensate for the inherent errors in low-performance gyroscopes. The resulting capability enables gyrocompassing, thereby eliminating the need for subsequent prolonged filtering phase (fine alignment). Through the development of theory and experimental validation, we demonstrate that the improved initial conditions establish a new lower error bound, bringing affordable gyros one step closer to being utilized in high-end tactical tasks.

Towards Learning-Based Gyrocompassing

TL;DR

This work tackles the challenge of gyrocompassing with low-cost MEMS gyroscopes by leveraging a learning-based approach to compensate inherent sensor errors during stationary self-alignment. A Bi-LSTM framework is proposed, paired with a cyclic mean squared error loss (CMSE) and data augmentation, to estimate the heading angle from gyro streams efficiently. Experimental results show the method can reduce the required alignment time by up to a factor of about 10 and lower heading error by more than 50% on unseen data, bringing practical gyrocompassing within reach for affordable sensors. The findings suggest that software-enabled gyrocompassing can extend high-end navigation capabilities to off-the-shelf instruments, with wide relevance for autonomous and tactical platforms.

Abstract

Inertial navigation systems (INS) are widely used in both manned and autonomous platforms. One of the most critical tasks prior to their operation is to accurately determine their initial alignment while stationary, as it forms the cornerstone for the entire INS operational trajectory. While low-performance accelerometers can easily determine roll and pitch angles (leveling), establishing the heading angle (gyrocompassing) with low-performance gyros proves to be a challenging task without additional sensors. This arises from the limited signal strength of Earth's rotation rate, often overridden by gyro noise itself. To circumvent this deficiency, in this study we present a practical deep learning framework to effectively compensate for the inherent errors in low-performance gyroscopes. The resulting capability enables gyrocompassing, thereby eliminating the need for subsequent prolonged filtering phase (fine alignment). Through the development of theory and experimental validation, we demonstrate that the improved initial conditions establish a new lower error bound, bringing affordable gyros one step closer to being utilized in high-end tactical tasks.
Paper Structure (22 sections, 22 equations, 11 figures, 4 tables, 2 algorithms)

This paper contains 22 sections, 22 equations, 11 figures, 4 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Literature Review
  3. Heuristic-based
  4. Filtering-based
  5. Optimization-based
  6. Theoretical Background
  7. Inertial sensors model
  8. Self-alignment model
  9. First-order error analysis
  10. Proposed methodology
  11. A learning approach
  12. Loss function
  13. Data augmentation
  14. Experimental Results
  15. Experimental setup
  16. ...and 7 more sections

Figures (11)

  • Figure 1: Diagrammatic representation of ECEF and NED frame.
  • Figure 2: Unrolled computational graph illustrating our Bi-directional LSTM model performing a many-to-one mapping.
  • Figure 3: Layout of the experimental setup.
  • Figure 4: An illustrative example of a 3D gyro measurements.
  • Figure 5: Allan deviation analysis for the gyroscope sensor data.
  • ...and 6 more figures