V3I7P1

Simulation and Evaluation of the Extended Kalman Filter in the Inertial Sensor Data Fusion Problem

Dao Thanh Bang^1* , Ngo Trung Toan¹

Abstract

This paper presents a method for simulating and evaluating the performance of the Extended Kalman Filter (EKF) in the inertial sensor data fusion problem, aimed at accurately estimating the three degrees of freedom (Roll, Pitch, Yaw) orientation states. The EKF algorithm is developed based on the nonlinear kinematic model of the system and is directly compared with measurement results from inertial sensors (gyroscope). Error parameters such as mean bias, root mean square error (RMSE), and variance are used for performance evaluation. Simulation results show that the EKF significantly reduces errors compared to the direct use of sensor data, while ensuring stability and reliability in the estimation process. The paper confirms the feasibility of applying EKF in inertial sensor data fusion for navigation and control applications.

Keywords:

Extended Kalman Filter; EKF; data fusion; inertial sensors; Roll-Pitch-Yaw; state estimation; simulation.