A Study of a GNSS/IMU System for Object Localization and Spatial Position Estimation

Area of Science:

• Robotics and Autonomous Systems
• Geomatics Engineering
• Sensor Fusion

Background:

• Navigation systems are crucial for autonomous vehicles, VR/AR, and object tracking.
• Accurate 3D position and orientation tracking are essential for these applications.
• Inertial Navigation Systems (INS) provide motion data but require fusion with other sensors for precise navigation.

Purpose of the Study:

• To integrate Global Navigation Satellite System (GNSS) with a 10 Degrees of Freedom (10DoF) Inertial Measurement Unit (IMU) system.
• To calculate object position, attitude, and heading accurately.
• To evaluate sensor data fusion techniques for improved navigation performance.

Main Methods:

• Sensor data fusion using two Kalman filters (KF) for position and attitude calculations.
• Detailed analysis of Allan Variance and normal distribution parameters for three MEMS IMU sensors.
• Performance evaluation of GNSS systems using commercial and proposed antennas.
• Experimental validation comparing KF heading angle output with other sources.

Main Results:

• The proposed sensor fusion approach enhances the accuracy of navigation tasks.
• Allan Variance and distribution analysis provide insights into MEMS IMU sensor characteristics.
• GNSS antenna performance is assessed for optimal system integration.
• Experimental results validate the effectiveness of the Kalman filter-based fusion for heading estimation.

Conclusions:

• The integration of GNSS and IMU systems with Kalman filter-based sensor fusion offers a robust solution for 3D navigation.
• Understanding sensor characteristics is vital for optimizing filter performance.
• The study demonstrates improved accuracy in position, attitude, and heading calculations for autonomous applications.