Real Time Rail Vehicle Running State Monitoring System using TinyML and IoT

The purpose of this research is to use data analysis using TinyML to build and implement a real-time rail vehicle running state monitoring system. In terms of data collection, processing speed, and dependability, this system's accuracy and performance are contrasted with those of conventional monitoring systems. Two groups are used for comparison in this study. Group 1: The suggested traditional rail monitoring system (Rule based or Conventional IOT) TinyML-based real-time railcar operating state monitoring system Our system's accuracy was 85.89% and its overall F1 score was 87.7%. Group 2: relates to railway transit for the purpose of tracking and controlling a number of factors proposed TinyML-based Rail Monitoring system including vehicle health, signalling, and track conditions. The range of data changes for various sensors was noted as a result, with Ultrasonic_Ir_Y ranging from 0.3 to 2.0 and DetectorX ranging from 0.01420, Lastly in this study, the proposed real-time rail vehicle running state monitoring system using TinyML and Data Analysis demonstrated significantly better performance compared to the traditional monitoring system. The TinyML-based system achieved higher accuracy, faster processing speeds, and more reliable real-time monitoring capabilities. This system effectively addresses the limitations of traditional systems, making it suitable for modern rail vehicle monitoring applications.