Improved attitude determination by compensation of gyroscopic drift by use of accelerometers and magnetometers

This paper presents a self-aiding scheme for improved attitude determination using low-cost MEMS-based inertial measurement unit consisting of three axis accelerometers, gyroscopes and magnetometers. The technique estimates and compensates gyroscope biases by use of sensors fusion mechanism. To achieve this, attitude is computed from gyroscopes through traditional rate integration scheme and the same is achieved from a combination of accelerometers and magnetometers through vector matching. The two attitudes computed are compared to form an attitude error to estimate biases of the gyros which are continuously adjusted through a feedback mechanism. This technique is compared with Kalman filter based data fusion algorithm which uses gyroscopes and a combination of accelerometers and magnetometers to estimates gyro biases. Both algorithms are tested on real data and showed comparable results.