Collective intelligence in adaptive IoT-based assistive technologies for visually impaired users in indoor environments

The increasing prevalence of visual impairment poses challenges to mobility and interaction, especially in indoor environments, where obstacles and accessibility barriers limit independence. While IoT and AI technologies offer promising solutions, many assistive systems lack real-time adaptability and do not fully meet the dynamic needs of visually impaired people (VIPs). This study explores MoSIoT (Modeling Scenarios of the Internet of Things), a monitoring system for people with disabilities, where an AI Orchestrator exemplifies collective intelligence by distributing decision-making across edge and cloud devices to improve indoor accessibility for VIPs. MoSIoT’s key features include object detection with distance measurement, audio-descriptive scene recognition, AI text reader, AI-powered chatbot, audio-descriptive color recognition, and adaptive IoT interaction control, providing a multimodal solution for navigation and device interaction. A system-level performance evaluation is conducted, measuring end-to-end latency across different execution paths and assessing the correctness of orchestration decisions. The Technology Acceptance Model (TAM) is used to assess initial user acceptance, focusing on perceived usefulness, perceived ease of use, and behavioral intention, along with factors such as experience, job relevance, and perceived external control. Data collected through structured questionnaires revealed that perceived usefulness and ease of use significantly influence adoption. Prior experience with assistive technology (AT), accessibility barriers, and environmental adaptability also play a critical role. This study presents a scalable, adaptive assistive framework for improving indoor accessibility for VIPs. The results provide initial evidence of user acceptance and highlight the importance of adaptive, real-time solutions based on collective intelligence to support independent navigation and interaction.