Drone-AI Mobile Sensing Intelligence for Smart X: A Multimodal Human-in-the-Loop Architecture

Abstract

Sensing is a foundational layer of Smart X systems because artificial intelligence can only support reliable decisions when environmental data are timely, contextual, and trustworthy. This paper proposes a Drone-AI Mobile Sensing Intelligence architecture that reframes unmanned aerial vehicles from flying cameras into multimodal mobile sensing platforms. The proposed artifact integrates Drone-IoT payloads, telemetry, adaptive connectivity, edge-cloud artificial intelligence, data governance, and human-in-the-loop decision control. The design is derived from a conceptual book chapter on Drone-AI sensing and is reformulated as an ICISS-style research paper through a design science orientation. The architecture is evaluated analytically through requirement mapping, a multimodal sensor taxonomy, a data-to-decision workflow, a conceptual PM2.5 monitoring scenario, a maturity model, and Smart X application mapping. The result is a reusable architecture for smart cities, smart environment, smart disaster, smart industry, smart mining, and smart campus contexts. The contribution is not a new drone hardware design, but a governance-aware sensing architecture that links mobile data capture, AI inference, forecasting, uncertainty handling, and accountable human action.