LoRa and IoT Based Landslide Early Detection System

Abstract

Safety is a top priority with the implementation of safety protocols, preparedness training, surveillance technology and early warning systems. A disaster is an event that causes loss, suffering and accidents in physical, economic, social and environmental forms. Common problems that often cause landslides are tree felling on slopes or due to unpredictable natural factors, resulting in landslides affecting settlements and causing fatalities. One solution to overcome this problem is to create an early detection tool for landslides with good and real-time communication to the internet. This tool is designed to facilitate monitoring of slopes that have the potential for landslides without having to go to the location directly. Research was conducted to determine the design and performance specifications of the tool. Safety is a top priority with the implementation of safety protocols, preparedness training, surveillance technology and early warning systems. A disaster is an event that causes loss, suffering and accidents in physical, economic, social and environmental forms. Common problems that often cause landslides are tree felling on slopes or due to unpredictable natural factors, resulting in landslides affecting settlements and causing fatalities. One solution to overcome this problem is to create an early detection tool for landslides with good and real-time communication to the internet. This tool is designed to facilitate monitoring of slopes that have the potential for landslides without having to go to the location directly. Research was conducted to determine the design and performance specifications of the tool. The research results are in the form of performance specifications and design specifications. Performance specifications consist of the manufacture of tool mechanics, tool electronic circuits, characteristics of the sliding potentiometer shift sensor and characteristics of the MPU6050 GY-25 tilt sensor. The results of sensor detection can be displayed on the serial monitor via the Arduino IDE application or on the Blynk application on Android. The design specifications for the LoRa and IoT-based landslide detection tool consist of two parts, namely the accuracy and precision of the landslide detection tool measurements with the following details: The average percentage of error in reading shift and slope values is 0.447% and 0.924% with an average accuracy of 98.147% and 97.252% respectively and an average accuracy of 97.251% and 99.553% respectively.