Constrained ESKF for UAV Positioning in Indoor Corridor Environment Based on IMU and WiFi.

Area of Science:

• Robotics and Automation
• Navigation Systems
• Sensor Fusion

Background:

• Indoor autonomous navigation for Unmanned Aerial Vehicles (UAVs) is challenging due to complex, unknown environments.
• Global Navigation Satellite System (GNSS) is unreliable indoors due to signal blockage and reflection.
• Accurate indoor positioning is crucial for UAV autonomy.

Purpose of the Study:

• To develop and evaluate an integrated WiFi and IMU positioning system for indoor UAV navigation.
• To improve positioning accuracy in indoor corridor environments.
• To demonstrate the effectiveness of sensor fusion for reliable indoor localization.

Main Methods:

• Utilized a zone partition-based Weighted K Nearest Neighbors (WKNN) algorithm for WiFi positioning.
• Implemented an Error-State Kalman Filter (ESKF) for fusing WiFi and IMU data.
• Applied a probability-based optimization method for further accuracy enhancement.

Main Results:

• Data fusion improved positioning accuracy by 51.09% over IMU-based and 66.16% over WiFi-based methods.
• Optimization further increased accuracy by 20.9% compared to the ESKF-based fusion.
• The integrated system demonstrated high indoor positioning accuracy using low-cost sensors.

Conclusions:

• The integration of WiFi and IMU provides a robust and accurate solution for indoor UAV positioning.
• The proposed methods overcome GNSS limitations in indoor environments.
• Low-cost sensor fusion is a viable approach for achieving precise indoor navigation.