Algorithmic model of the technological process for spatial navigation of blind and visually impaired persons based on predictive trajectory correction

Modern navigation systems using AI and computer vision remain poorly adapted for blind and visually impaired users, as existing solutions often lack predictive capabilities and adaptability in dynamic environments. To address these limitations, this study aims to develop an algorithmic model of spatial navigation that integrates predictive trajectory correction and quantitative threat assessment for both dynamic and static obstacles.

The proposed model structures the navigation cycle into scene formation, object recognition, motion prediction, and route adjustment, introducing a hazard coefficient k to formalize risks, scale danger zones, and adapt system behavior in real time. Its practical significance lies in its potential to serve as the foundation for intelligent assistive devices, including mobile applications, wearable systems, smart canes, and robotic assistants, enhancing user safety, reducing cognitive load, and providing adaptive threat warnings.