Surakshini: A Self-Powered IoT-Based Anomaly Detection System for Railway Track Safety Using TinyML and LoRa

Abstract

Railways form the backbone of logistics and connectivity in India, but ensuring the safety of trackbeds in remote sections remains a growing challenge. This paper proposes “Surakshini,” a self-powered IoT-based anomaly detection system using TinyML (Tiny Machine Learning) and LoRa (Long Range). Surakshini detects vibrations and identifies potential anomalies in track behavior using a local machine learning model. The system operates autonomously with solar power and transmits alerts over long-range LoRa communication. Our prototype demonstrates accurate detection of train approach and rail condition issues with minimal power consumption. The proposed solution enhances railway safety infrastructure using edge intelligence and sustainable design.