e-Rshad: Enabling Inclusive Indoor Navigation for Individuals with Visual Impairments (CMUQ - Seed)

Eduardo Feo Flushing (LPI), Khaled Harras, Mai Al-Mashhadani, Yaqoob Ansari, Gustavo Ramalho, Abdulrahman Al-Taweel, Yousef Hussein

This project aims to develop e-Rshad, a conversational and adaptive navigation system designed for individuals with visual impairments. Its purpose is to enhance their navigation experience in complex public spaces such as shopping malls, museums, and airports. Traditional navigation tools often fall short in these areas due to their high cost, limited availability, and reliance on privacy-compromising cameras. This proposal aims to bridge these gaps by using alternative privacy-preserving sensor modalities like low-cost mmWave radar and low-resolution thermal sensors, along with recent advances in AI, to ensure reliable navigation. The project will also create comprehensive datasets tailored to assistive navigation, leading to advancements in the field. The contributions of this project will empower individuals with visual impairments, fostering their independence and improving their quality of life through a more accessible, privacy-aware indoor navigation solution.

Related Publications

Directional Navigability Estimation from Ultra-Low-Resolution Multi-View Thermal Sensing

Tesseract: Interactive Conversion of Building Floor Plans into Multi-Floor Navigable Graphs [Demo]

GraphWalker: Deployable Inertial-Based Indoor Localization via Trajectory to Graph Alignment

Towards Practical and Deployable Infrastructure-Free Indoor Localization

Tesseract: Unfolding Navigable Graph Representations from Low-Semantic Floor Plans

A Deployable Privacy-Preserving Thermal-Based Obstacle Detection System for Indoor Navigation

Text2Map: From Navigational Instructions to Graph-Based Indoor Map Representations Using LLMs