A Map/INS/Wi-Fi Integrated System for Indoor Location-Based Service Applications.

Area of Science:

• Robotics
• Geomatics
• Computer Science

Background:

• Indoor location-based services (LBS) face challenges with accuracy and continuity.
• Inertial Measurement Units (IMUs) suffer from cumulative drift errors.
• Integrating diverse sensor data is crucial for robust positioning.

Purpose of the Study:

• To develop an accurate, continuous, and low-cost indoor positioning system.
• To effectively fuse data from maps, INS, and Wi-Fi signals.
• To overcome the limitations of standalone indoor positioning methods.

Main Methods:

• A cascaded Extended Kalman Filter (EKF) and Auxiliary Value Particle Filter (AVPF) framework was employed.
• Wi-Fi signals and non-holonomic constraints corrected IMU drift via EKF.
• Map-aiding and map-matching methods constrained Wi-Fi/IMU positions using AVPF.

Main Results:

• The integrated system significantly reduced positioning errors compared to standalone INS.
• The proposed method demonstrated stable, continuous, and reliable indoor location estimation.
• Effective fusion of multi-source information was achieved through the cascaded filter structure.

Conclusions:

• The Map/INS/Wi-Fi integrated system offers a viable solution for accurate indoor LBS applications.
• The cascaded EKF/AVPF filter effectively mitigates sensor drift and enhances positioning performance.
• This approach provides a robust and computationally efficient method for indoor navigation.