Attitude Estimation and Control Using Linear-Like Complementary Filters: Theory and Experiment
This paper proposes new algorithms for attitude estimation and control based on fused inertial vector measurements using linear complementary filters principle. First, n-order direct and passive complementary filters combined with TRIAD algorithm are proposed to give attitude estimation solutions. These solutions which are efficient with respect to noise include the gyro bias estimation. Thereafter, the same principle of data fusion is used to address the problem of attitude tracking based on inertial vector measurements. Thus, instead of using noisy raw measurements in the control law a new solution of control that includes a linear-like complementary filter to deal with the noise is proposed. The stability analysis of the tracking error dynamics based on LaSalle's invariance theorem proved that almost all trajectories converge asymptotically to the desired equilibrium. Experimental results, obtained with DIY Quad equipped with the APM2.6 auto-pilot, show the effectiveness and the performance of the proposed solutions.