Attitude Estimation for Rigid Aircraft with Time-Varying Gyro Bias: A Finite-Time Complementary Filtering Approach

This paper proposes a novel finite-time complementary filtering approach for attitude estimation of rigid-body aircraft subject to time-varying gyro bias. In contrast to conventional complementary filters that only ensure asymptotic convergence, which merely guarantee asymptotic convergence, the proposed filter guarantees rapid convergence to the actual attitude within a finite time, while effectively compensating for dynamic gyro bias. The filter’s adaptability to time-varying biases significantly enhances the accuracy and robustness of attitude estimation in dynamic operating conditions. The finite-time convergence of the proposed filter is rigorously established through Lyapunov function-based stability analysis, with the settling time explicitly characterized in terms of the filter parameters and initial estimation errors. Finally, numerical simulations are conducted to validate the superior performance of the filter, demonstrating faster convergence, improved accuracy, and strong robustness against measurement disturbances.