FPP-SLAM: indoor simultaneous localization and mapping based on fringe projection profilometry

Area of Science:

• Robotics and Artificial Intelligence
• Computer Vision and 3D Sensing

Background:

• Simultaneous Localization and Mapping (SLAM) is crucial for autonomous systems but faces challenges indoors.
• Existing indoor SLAM methods using LiDAR or GNSS have limitations in accuracy and signal availability.
• Fringe Projection Profilometry (FPP) is a promising 3D sensing technique with underutilized potential for SLAM.

Purpose of the Study:

• To introduce Fringe Projection Profilometry (FPP) as a viable sensor for indoor Simultaneous Localization and Mapping (SLAM).
• To develop and propose novel SLAM algorithms tailored for FPP in indoor environments.
• To enable millimeter-level, real-time indoor mapping and localization using FPP-SLAM.

Main Methods:

• Integration of FPP 3D sensing technology into an indoor SLAM framework.
• Development of specialized SLAM algorithms optimized for FPP data.
• Real-time data processing for simultaneous mapping and localization.

Main Results:

• The proposed FPP-SLAM system achieves millimeter-level accuracy in indoor environments.
• Real-time performance demonstrated in both simulated and real-world room-sized scenes.
• Outperformed state-of-the-art methods in terms of mapping and localization efficiency and accuracy.

Conclusions:

• FPP is effectively introduced as a robust sensor for indoor SLAM applications.
• The developed FPP-SLAM algorithms provide a high-accuracy, cost-effective solution for indoor navigation.
• This work establishes a foundation for the broader adoption of FPP in indoor SLAM.