ABSTRACT

This chapter presents various existing techniques of state estimation for a nonlinear continuous-discrete system. In many real life problems, process models are continuous because they are derived from the law of physics and a measurement equation is discrete because sensor outputs are sampled. In many physical systems, the process is described by a continuous time stochastic differential equation and measurements are also received continuously. In many analog control systems which operate without digital computers, the measured signals are in the continuous time domain. The chapter discusses continuous time filtering where the process and measurement equations are both in a continuous time domain. It describes the continuous-discrete system where the process model is described in continuous time but the measurements remain in a discrete time domain. The derivation of the continuous-discrete Kalman filter (KF) is a little different from the continuous time KF because measurement is available only at discrete time-steps.