Abstract

This work presents a novel, drift-resistant hybrid navigation system designed to provide robust, real-time localization and mapping in environments where Global Positioning System (GPS) signals are unavailable or compromised. The study focuses exclusively on the synergistic fusion of Stereo Vision Simultaneous Localization and Mapping (S-SLAM) and Real-Time Location Systems (RTLS). The S-SLAM component delivers high-resolution, local 3D geometric awareness and relative pose estimation by utilizing real-time disparity mapping and visual feature tracking. However, to mitigate the inherent drift associated with vision-based dead reckoning, the system incorporates absolute position fixes from an established RTLS infrastructure (e.g., UWB). The fusion architecture tightly integrates the local relative pose data from the stereo cameras with the global, non-cumulative position updates from the RTLS. This combination yields highly stable and accurate position and orientation estimates for mobile platforms, creating a powerful and practical solution for demanding operational contexts, including tunnels, mines, dense urban canyons, and industrial facilities.

About the Speaker

Ms. Wissem Dilmi is a researcher and specialist focused on advanced control and localization solutions for robotic systems. She holds a master’s degree in Control and Instrumentation from Algeria, obtained in 2021. Her research interests are centered on robust, embedded autonomy, particularly the design and implementation of highly accurate navigation systems. This includes expertise in the Hybrid Vision based Control and Quantum Navigation with Real Time Location Systems to ensure stable positional integrity in environments where GPS signals are absent or degraded.

Ms. Wissem Dilmi has hands-on experience in robotic systems, embedded control, and hardware experimentation. She also holds a professional role as a Project Manager in industry, leveraging her technical background to deliver complex projects.