This paper presents a study on the use of the Madgwick filter to normalize signals from Inertial Measurement Units (IMUs). IMUs are widely used systems for measuring and tracking the motion of objects, but the signals from IMU sensors are often corrupted by bias and accumulated noise, which reduces the measurement accuracy. The Madgwick filter is a complementary filter that combines data from accelerometers, gyroscopes, and magnetometers to estimate the orientation of the sensor. The purpose of this study is to evaluate the effectiveness of the Madgwick filter in improving the measurement accuracy of IMU signals. The results show that the Madgwick filter significantly reduces noise and errors in IMU signals while improving the measurement accuracy. This article provides a detailed description of the Madgwick filter and its implementation, which can be applied to human motion analysis systems, robot navigation, and unmanned aerial vehicle control.