A co-operative control approach to the regulation of nonlinear discrete-time structured systems

This work extends previous research dealing with the deadbeat out put regulation of discrete-time nonlinear plants. The plant is composed of agents that interact, via scalar-valued signals, in a known structured way. As in our earlier research, control agents are introduced to interact with certain plant agents, where each control agent focuses on regulating a specific plant agent, called its target. Targeting analysis is used to determine if control laws can be found to regulate all target agents, then growing analysis is used to determine the effect of those control laws on non-target plant agents. Regulation is achieved if these analyses indicate that all plant agents can be regulated. This paper goes beyond previous work by extending the class of allowable plant dynamics. Moreover, new necessary and sufficient algebraic conditions are derived to determine when targeting succeeds, and new easily-verifiable conditions necessary for targeting and/or growing to succeed are presented. These new conditions concern the positioning of control agents and targets as well as the propagation time of signals through the plant.