V3I6P21

Integration of Multi-Frequency Oscillation Models and Pseudo Linear Acceleration for Advanced IMU Sensor Data Simulation and Attitude Estimation Accuracy Enhancement 

Ngo Trung Toan^1*

Abstract

This paper presents a method of simulating IMU data for ships by combining a multi-frequency wave model with pseudo linear acceleration to evaluate and improve Euler angle estimation. While previous studies only considered ship oscillations based on the multi-frequency wave model, we introduce the linear acceleration component generated from ship body motions (surge, sway, heave), which is unrelated to gravity but appears in accelerometer data, causing significant errors in angle estimation algorithms. The simulated data include both ideal signals and noisy signals, enabling performance comparison between gyro-only and Complementary Filter in estimating Roll, Pitch and Yaw. The results show that adding pseudo linear acceleration makes the simulation closer to real-world conditions and provides a basis for evaluating IMU data processing algorithms under pseudo-noise conditions.

Keywords:

IMU simulation, Euler angle, multi-frequency wave model, pseudo linear acceleration, complementary filter, sensor bias.