A New Complementary Filter Design for Improved Performance

Abstract

When using a single gyroscope sensor for posture estimation, integrating the sensor data can lead to errors due to drift. Similarly, using only accelerometer for measurements results in significant noise problems. To address these problems, a solution exists in the form of a complementary filter that combines the strengths of both sensors while mitigating their drawbacks. In this paper, we study a new complementary filter that can effectively remove the effect of sensor noise. By modifying the structures of low pass filter and high pass filter, a new complementary filter is presented. Through simulation and experimental studies, we verify the effectiveness of the proposed approach.