A Compensation Controller for Stochastic Nonlinear Systems

Abstract

The invention discloses a compensation controller applied to a random nonlinear system, which belongs to the field of automatic control. The controller comprises a control input unit which is used for inputting a control input amount, a control output unit which is used for outputting a system output amount, an error input unit which is used for receiving an external interference amount, a compensation unit which is used for calculating and storing the control system state, a system construction unit which is used for constructing a discrete-time non-linear system affected by state-dependent noise, a constraint condition input unit which is used for inputting a space for a quadric surface constraint condition, a matrix decomposition unit which is used for matrix decomposition of the quadric surface constraint condition, and a data packet loss model unit which is used for characterizing that a random variable for the probability of state measurement data loss conforms to a Bernoulli process. State data are successfully transferred through a communication network.

Description

technical field [0001] The invention relates to a compensation controller for a random nonlinear system in the field of automatic control. Background technique [0002] In the past few decades, stochastic systems have been widely used in various aspects, and the research on the robustness and stability of stochastic systems has made great progress. A current hotspot for stochastic systems is the finite-time stability for linear stochastic systems. The stability of a linear stochastic system in finite time means that within a given finite time period, all permissible input disturbances will not cause its output to exceed a given upper threshold. With the rapid development of communication networks, more and more information is transmitted through communication networks. Some problems will occur after introducing the communication network into the feedback control loop, where the packet loss of the measurement data is considered as the main factor for the performance deterio...

AI Technical Summary

Problems solved by technology

Some problems occur after introducing the communication network into the feedback control loop, where the packet loss of the measurement data is considered as the main factor for the performance deterioration of the control system

At present, there is almost no research on the finite-time stability of random systems with packet loss of measurement data. Simultaneously considering the stability of nonlinear stochastic systems with packet loss of measurement data in finite time will make the problem more complicated.

Images