ABSTRACT
Scattered and dense human population adjacent to the railway line, frequent blind turns on the track, unmanned level crossing, reduction in visibility due to extreme climatic conditions (rain, fog etc.), and reckless track crossing by a section of native inhabitants are coursing frequent rail accidents in India. To counter these GPS tracking, machine monitoring, and several highend technologies have surfaced over the years but high cost and technological complexity have limited their overextended use. Moreover, those accidents are local hence a less complex ultra-low-cost location-wise system is a more desired solution. To address the gap, an ultra-low-cost indigenous Real-time Train Alert System (RTAS) is designed. The RTAS works with two standalone vibration sensors which are fixed on the rail track. The sensors capture vibrations from approaching locomotive from hundreds of meters away and the signal source authentication is done with an innovative algorithm and aboriginal hardware unit. Once detected and authenticated, the RTAS will switch on a high-volume hooter and flashlight to warn fellow citizens about approaching train and thus will avoid potential accidents. Under present work system design, algorithm, etc. are presented in detail and system performance is verified through simulated experiments on a software-hardware integrated platform.
