Stochastic Stable Attitude Estimation Algorithm Using UKF With Measurement Loss

Measurement loss and initial attitude errors caused by the sensor’s complex information transmission process and environmental interference are challenging topics in spacecraft attitude estimation. In this article, a new class of unscented quaternion estimator based on loss matrix with measurement loss is proposed to deal with this problem. The attitude estimation system model with measurement loss is established using the Bernoulli random variable. The measurement update process of the proposed algorithms is redesigned by introducing the expected value of the measurement loss matrix into the filtering recursion process. The boundedness of the error covariance and the stochastic stability of the estimated state are then theoretically verified. According to the stability analysis, a scheme for adding adjustment factors is proposed. The simulation results show that the proposed algorithms could achieve better performance compared with the state-of-the-art algorithms in terms of convergence rate, estimation accuracy, and stability.