Improving the Accuracy of Decawave's UWB MDEK1001 Location System by Gaining Access to Multiple Ranges

Area of Science:

• Robotics and Automation
• Wireless Communication and Networking
• Indoor Positioning Systems

Background:

• Accurate indoor localization is crucial for people, robots, and IoT devices.
• Ultrawideband (UWB) offers centimeter-level accuracy but is degraded by indoor obstacles.
• Existing UWB systems face hardware limitations, restricting range measurements and redundancy.

Purpose of the Study:

• To address the limitations of commercial UWB systems regarding measurement redundancy.
• To develop and demonstrate a procedure for increasing the number of range measurements per tag.
• To improve the robustness and accuracy of UWB-based indoor localization.

Main Methods:

• Implemented a strategy involving duplicated mobile tags and a double interleaved anchor network.
• Utilized a wireless BLE protocol via an API to obtain up to eight range measurements per tag at 10 Hz.
• Designed a robustified Extended Kalman Filter (EKF) for position estimation using trilateration and outlier removal.

Main Results:

• Successfully enabled each mobile tag to measure ranges to eight anchors, overcoming the commercial system's four-anchor limit.
• Demonstrated improved position estimation accuracy compared to the standard four-range commercial solution.
• Validated the multi-range approach in both laboratory and apartment environments.

Conclusions:

• Increasing range measurement redundancy significantly enhances UWB localization accuracy in indoor environments.
• The proposed method effectively mitigates the impact of signal obstructions and outliers.
• This multi-range UWB approach offers a practical solution for more reliable and precise indoor positioning.