Sistem Persinyalan Kereta Api Cerdas Dengan Sensor Huskylens Dan Kendali Otomatis Berbasis Internet of Things

Abstract

One of the safety factors in the world of railways is a reliable and excellent signaling system. Railway signaling consists of two components, namely electronic and mechanical parts, in the electronic part there are light signals with red, yellow and green indicators like traffic lights on the highway, for the mechanical part there is a wesel motor component that functions as a regulator of the direction of turn or rate of the train to be addressed, and for the signaling system used is the interlock system. Interlock is a relay-based device that functions to secure train travel and minimize train accidents. In 2009 there was an accident between Senja Utama Semarang train and Argo Bromo Anggrek train due to power outage and driver's negligence in reading signal lights. This research aims to design IoT-based automatic signaling on trains and automate train control to reduce the number of accidents. The train that will be used is a miniature train with HO scale 1:87 with a layout of 2 station lines and 1 non-station line, a servo motor is used as a turn signal, an IR sensor (FC-51) as a train detector and as a previous signal light colour changer, and a HUSKYLENS camera sensor as a signal light colour detector on an automatic train. In the interlock system using the Finite State Machine (FSM) method applied on the Arduino Mega, this system is monitored via a website. The train control system is applied on ESP32 and controlled through Blynk. The design of the signaling system and automatic train control has been able to achieve a 100% success rate with the implementation of an interlocking system with an average delivery time from ESP32 to the server is 2.779 seconds. Light signals, switches, and position sensors can be monitored directly through the website. Both automatic and manual train control can be controlled directly through the Blynk application.