Enhancing Indoor Positioning with Acoustic Signal-Aided Visual-Inertial SLAM

To address the issue of long-term trajectory drift caused by cumulative errors in visual-inertial SLAM for indoor positioning scenarios, we propose a method that augments visual-inertial SLAM with acoustic signal. Specifically, we employ factor graph optimization to incorporate acoustic ranging and relative velocity measurements as global factors into the SLAM system for joint optimization, aiming to reduce cumulative errors and improve accuracy. Benefiting from the use of TOA and FOA positioning methods in the acoustic component, our proposed method is of significant importance for indoor scenarios with obstructions and low beacon density. We validate our positioning method by simulating with acoustic measurements on the public dataset EuRoC. The experimental results demonstrate that compared to pure visual-inertial SLAM, our method achieves superior performance in positioning accuracy, proving the effectiveness of this approach in reducing long-term trajectory drift.