On rate allocation for multiple plants in a networked control system

• The problem of allocating communication resources to multiple plants in a networked control system is investigated. In the presence of a shared communication medium, a total transmission rate constraint is imposed. For the purpose of optimizing the rate allocation to the plants over a finite horizon, two objective functions are considered. The first one is a single-objective function, and the second one is a multi-objective function. Because of the difficulty to derive the closed-form expression of these functions, which depend on the instantaneous communication rate, an approximation is proposed by using high-rate quantization theory. It is shown that the approximate objective functions are convex in the region of interest both in the scalar case and in the multi-objective case. This allows to establish a linear control policy given by the classical linear quadratic Gaussian theory as function of the channel. Based on this result, a new complex relation between the control performance and the channel error probability is characterized.

Nyckelord

Channel error probabilities

Closed-form expression

Communication rate

Communication resources

Control performance

Finite horizons

High-rate quantization

Linear controls

Linear quadratic Gaussian theory

Multi objective

Multi-objective functions

Objective functions

Rate allocation

Region of interest

Shared communication medium

Total transmission rate

Linear control systems

Networked control systems

Communication

Publikations- och innehållstyp

ref (ämneskategori)

kon (ämneskategori)