An Extensible Positioning System for Locating Mobile Robots in Unfamiliar Environments

Area of Science:

• Robotics
• Navigation Systems
• Sensor Fusion

Background:

• Accurate mobile robot localization is crucial for autonomous operation, especially in unknown or dynamic environments.
• Existing systems often require complex infrastructure, including inter-anchor communication and clock synchronization.

Purpose of the Study:

• To propose an extensible positioning system for mobile robots that does not require inter-anchor communication or clock synchronization.
• To enable robust localization in unfamiliar environments by allowing system extension with new anchors.

Main Methods:

• A system combining stereo cameras, an inertial measurement unit (IMU), and an ultra-wideband (UWB) network.
• Recursive least squares (RLS) for estimating new anchor positions.
• Maximum correntropy Kalman filter (MCKF) for fusing UWB and IMU data.
• Visual SLAM and UWB system comparison for initial attitude calculation.

Main Results:

• Achieved a root mean square error (RMSE) of 0.130 m with all anchors.
• Demonstrated robust performance in unfamiliar environments with an RMSE of 0.131 m.
• New anchor position estimation achieved an Euler distance RMSE of 0.061 m.

Conclusions:

• The proposed extensible positioning system offers high accuracy and adaptability for mobile robots.
• The system effectively localizes robots in unknown environments by seamlessly integrating new UWB anchors.
• The method overcomes limitations of traditional UWB systems by removing communication and synchronization requirements.