Edge AI for Real-Time Decision Making

Abstract

Edge Artificial Intelligence (Edge AI) is rapidly transforming the landscape of real-time decision-making across various domains. Unlike traditional cloudbased AI systems, Edge AI brings computation closer to the data source, reducing latency, bandwidth usage, and improving privacy. This review paper discusses the key concepts, architectures, and applications of Edge AI, focusing on real-time decision-making scenarios. The paper highlights recent developments in hardware accelerators, software frameworks, and algorithmic optimizations that enable AI models to operate efficiently on edge devices. Challenges, including energy constraints, model accuracy, and security concerns, are discussed along with potential solutions. Future directions and emerging trends in Edge AI are also explored to guide researchers and practitioners in this growing field.

KEYWORDS: Edge AI, real-time decision making, IoT, AI on edge devices, low-latency AI, embedded intelligence, edge computing.