Title: AI, Innovation and Global Transformation: Interdisciplinary Perspectives on Technology, Business and Society
Editors: Dr. J. Preetha, and Dr. Siddhartha Mehrotra
ISBN: 978-81-69857-64-2
Chapter: 29
DOI: https://doi.org/10.59646/809/29
Author: R. Kokila Devi
Abstract
Ambient Intelligence (AmI) and invisible computing represent a paradigm shift where computational assistance integrates seamlessly into physical architecture, operating autonomously without explicit user intervention. Conventional smart environments frequently rely on intrusive visual sensors, fragile cloud-tethered pipelines, and high-latency polling routines, compromising user privacy and causing operational bottlenecks. This chapter proposes an end-to-end, edge-native, zero-interaction ambient intelligence framework powered by non-intrusive multi-modal sensing. By combining radio-frequency micro-Doppler radar signatures, ultrasonic spatial ranging, environmental telemetry, and sub-surface piezoelectric floor vibrational dynamics, the architecture maps human indoor activity, spatial occupancy, and behavioral anomalies. A Spatio-Temporal Graph Neural Network coupled with dynamic contextual rule arbitration executes local real-time inference on low-power edge nodes. Comprehensive empirical evaluations demonstrate high activity recognition accuracy, sub-50-millisecond closed-loop actuation latency, and low power consumption, establishing a practical model for private, resilient, and unobtrusive ambient intelligent environments across assistive eldercare, enterprise workstations, and smart residential spaces.
Keywords: Ambient Intelligence; Invisible Computing; Edge Intelligence; Non-Intrusive Sensing; Graph Neural Networks; Context-Aware Computing.