Browsing by Author "Azfar, Samnun"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    A Data Driven Toggling Gain Complementary Filtering Approach for Orientation Estimation
    (Department of Computer Science and Engineering(CSE), Islamic University of Technology(IUT), Board Bazar, Gazipur-1704, Bangladesh, 2025-10-25) Azfar, Samnun; Audhi, Ramisa Zaman; MdInzamam, Mir
    One of the goals of an Inertial Navigation System (INS) is to estimate the 3D orienta tion of a bodyundermotionbasedon givenaccelerometer,magnetometerandangular rate gyroscope readings. Complementary fusion is one of the most robust, mathe matically simple and fast algorithms for fusing the accelerometer, magnetometer and angular rate gyroscope readings to estimate the 3D orientation. But complementary fusion of sensor data suffers from linear accelerations and constant gain problems. This thesis aims to solve the aforementioned issues in a data driven approach. In this work, we introduce two contributions for robust orientation estimation using raw in ertial and magnetic sensor data. First, we propose a tree-based Extreme Gradient Boosting (XGB) model that effectively denoises raw gyroscope, accelerometer, and magnetometer signals. Second, we present a tree-based movement detection model that dynamicallyadaptsfusionprocessthrough𝛼-toggling,enablingimprovedrobust ness to linear accelerations and magnetic interference. Together, these contributions establish a data-driven fusion framework that enhances reliability beyond traditional fixed-gain filters. The proposed approach is validated on a real-world dataset and fur ther evaluated using a custom-built mechanical turntable, providing controlled test ing conditions for systematic performance assessment