Analysis and Compensation of Installation Errors for Rotating Semi-Strapdown Inertial Navigation System

Jiayu Zhang, Jie Li, Yugang Huang, Chenjun Hu, Kaiqiang Feng, Xiaokai WeiIEEE Access 2019 · DOI

Abstract

The rotation semi-strapdown inertial navigation system (RSSINS) provides a new solution for the high-precision measurement of high-spinning projectile flight parameters. However, the installation error between the complex semi-strapdown mechanical structures is inevitable. Coupled with the carrier motion, the system installation error seriously affects the navigation accuracy of the system, which should be accurately calibrated and compensated. Aiming at the problem, the calibration and compensation methods are proposed in this study. First, the generation mechanism of the installation errors and the error propagation equation are analyzed and derived. Furthermore, the influence of each installation error angle on attitude is analyzed by simulation. Based on the analysis, taking MEMS device errors into account, a graded calibration method suitable for the system is proposed. In addition, according to the characteristics of the calibration method, the corresponding compensation method is proposed. Finally, experiments have been carried out to validate the performance of the method, which shows that the proposed method can compensate the installation errors effectively, and the measurement accuracy of the RSSINS has been improved.

The High-Spin Challenge

High-spinning projectiles, such as artillery shells, present a significant challenge for standard MEMS-based Strapdown Inertial Navigation Systems (SINS). The extreme roll rates often exceed the dynamic range of high-precision MEMS gyroscopes. To mitigate this, engineers utilize a Semi-Strapdown approach: sensors are housed within an inner cylinder that is motor-driven to rotate in the opposite direction of the projectile. This effectively "cancels out" the high spin, allowing the sensors to operate in a low-dynamic environment.

Mechanical structure of the Rotating Semi-Strapdown Inertial Navigation System (RSSINS).

However, this mechanical complexity introduces a critical problem: installation errors. It is physically impossible to perfectly align the axes of the inner cylinder, the motor, and the outer projectile body. These tiny misalignments couple with the projectile's motion to create massive navigation drifts. Because the system has nested moving parts, traditional "all-in-one" calibration methods cannot distinguish which specific component is misaligned.

Modeling Installation Errors

The researchers derived a mathematical model using Direction Cosine Matrices (DCM) to describe how misalignments in the inner cylinder (sensor-to-motor) and outer cylinder (motor-to-carrier) propagate into attitude errors.

Schematic of the misalignment between the inner and outer cylinder axes.

A key insight from their simulation analysis was that outer cylinder errors are far more destructive because they couple directly with the projectile's high roll rate. While inner cylinder errors are coupled with the relatively slow modulation motor speed, outer errors create rapidly oscillating errors that bleed into pitch and yaw measurements.

Graded Calibration and Compensation

The authors propose a graded calibration and compensation method to decouple these errors:

  1. Phase 1 (Inner Isolation): The inner cylinder is rotated forward and backward at specific rates while the rest of the system remains stationary. This isolates the inner cylinder's specific installation angles.
  2. Phase 2 (System Integration): The entire system is placed on a tri-axis turntable. By comparing the known turntable motion to the system's output during specific rotations, the remaining coupled error terms are solved.
  3. Real-time Compensation: Rather than attempting physical hardware adjustments, the calculated error angles are used to "un-distort" the navigation math in real-time by expanding the direction cosine matrix used in the navigation solution.

Experimental Results

The study validated the method using a physical RSSINS prototype mounted on a high-precision tri-axial flight simulator to replicate high-spin conditions. The proposed compensation method showed marked improvements in attitude accuracy:

  • Pitch Error: Reduced from $-1.515^\circ$ to $-0.769^\circ$.
  • Yaw Error: Reduced from $1.237^\circ$ to $0.5453^\circ$.
  • Roll Error: Reduced from $-2.386^\circ$ to $-1.778^\circ$.

While roll error remains the highest (as it is the axis of modulation), the accuracy for pitch and yaw—critical for guidance—was brought well under the $1^\circ$ threshold.

Conclusion and Future Work

The graded approach effectively isolates errors that are otherwise unobservable in a single-step calibration. However, limitations remain. The current study focuses on short-endurance ballistic flights (approx. 60 seconds); for longer-range applications, residual errors might still lead to significant position drift. Additionally, future iterations could integrate thermal compensation to account for mechanical shifts caused by the extreme heat generated during high-speed flight.

Summary updated

BibTeX
@article{zhang2019analysis,
  author = {Jiayu Zhang and Jie Li and Yugang Huang and Chenjun Hu and Kaiqiang Feng and Xiaokai Wei},
  title = {{Analysis and Compensation of Installation Errors for Rotating Semi-Strapdown Inertial Navigation System}},
  journal = {IEEE Access},
  year = {2019},
  doi = {10.1109/access.2019.2930988},
}

Loading related content...

Figure