Robot Position Measurement Using Inertial Measurement Unit and Random Forest

This research discusses an innovative approach to measuring robot position using Inertial Measurement Unit (IMU), sensor fusion, and Random Forest. In this research, synthetically generated sensor data, including angular velocity and acceleration, is used to train a Random Forest Regressor-based machine learning model. The sensor fusion process occurs through combining information from the IMU, which includes angular velocity and acceleration data. The resulting model is then evaluated using the Mean Squared Error metric and the coefficient of determination. The experimental results demonstrate the potential of this approach in accurately estimating robot position by utilizing IMU sensors and artificial intelligence techniques. This study illustrates the interplay between sensor data, information fusion, and machine learning to improve robot position measurement capabilities. From the method in this study, it shows quite good accuracy with an r squared value of 0.9524860132061431