Secure State Estimation for Nonlinear Systems Under Sparse Attacks With Application to Robotic Manipulators

Abstract

Secure state estimation against sparse injection attacks and disturbances is a challenging problem of both theoretical and practical importance, and existing results mainly focus on linear systems despite many practical systems being nonlinear. In this paper, a novel secure state estimation scheme is proposed for a class of nonlinear systems with application to robotic manipulators. A kind of high-gain K-filters is constructed to estimate the unmeasured states, which can attenuate the disturbances to an arbitrary level. Moreover, a monitoring function and a switching scheme are introduced, which successfully preclude attacked measurements after a finite number of switchings. With these efforts, the proposed estimation scheme steers the estimation error into a residual set which can be made arbitrarily small by properly choosing some design parameters, regardless of the disturbances and possibly unbounded attacks. Both simulation and experimental results on a robotic manipulator demonstrate the effectiveness of the proposed method.