Attitude control of rigid body dynamics from biased IMU measurements

Commercially viable aerial robotic vehicles require low-cost attitude stabilisation systems that are robust to noise and sensor bias. A typical attitude stabilisation system consists of MEMs accelerometers, gyroscopes linked to separate attitude estimator and attitude controller algorithms. This paper proposes a non-linear attitude stabiliser for low-cost aerial robotic vehicles that combines attitude and bias estimation with control. The attitude control algorithm is based on a non-linear control Lyapunov function analysis derived directly in terms of the rigid-body attitude dynamics and measurement signals.