Evaluation and calibration of MEMS-IMU sensors for real-time landslide detection and monitoring system

Abstract Landslide detection and monitoring are crucial for the development of effective early warning systems and risk mitigation strategies. Recent advancements in Micro-Electro-Mechanical Systems (MEMS)-based Inertial Measurement Units (IMUs) technology have emerged as a powerful tool for landslide detection and early warning. This study compares and calibrates MEMS-based IMU sensors for real-time landslide detection and monitoring systems. Three IMU devices-Pixhawk, MPU6050, and BNO055-were evaluated through a series of controlled laboratory experiments designed to assess their performance in terms of noise levels, accuracy, and consistency of the readings. The results showed that the BNO055 sensor exhibited superior performance to the other sensors, with substantially lower noise levels and greater accuracy in measuring gravitational acceleration at various inclinations under static conditions. The sensor’s built-in fusion algorithm further enhances its ability to separate linear and gravitational accelerations, providing comprehensive data to differentiate slope movements and identify failure types. Additionally, the constant rotation experiment verified the BNO055’s gyroscope accuracy, confirming its suitability for real-world applications. The two-point flip test’s calibration for the offset and sensitivity mismatch effectively reduced the sensor’s error margin to within 0.5 degrees for typical slope inclinations. These findings establish the BNO055 sensor as the most suitable choice for detecting real-time slope movements, contributing to the effectiveness of monitoring efforts and the development of robust landslide detection and early warning systems.