Search

This paper focuses on the behavior of a linear quadratic Gaussian (LQG) controller for discrete-time systems utilizing industrial communication protocols. A new design of LQG problem is investigated for constrained networked control systems by using a new class of a quadratic cost function with a communication cost. This paper is considered two industrial communication protocols namely: the transportation control protocol (TCP), in which the sender should receive an acknowledgment regarding the delivery of the sending packets and the user datagram protocol (UDP) where the acknowledgment signal is absence. In the latter one, the driver (actuation) does not afford acknowledgment to inform the controller/estimator regarding the reception of the action packets. In such setting, the UDP-like protocol can react nonlinearly, where the absence of the acknowledgment creates a nonlinear optimization control law. In this sense, a LQG with standard Kalman filtering is failed because the separation principle can not reached. Under this setting, a suboptimal methodology can be successfully implemented to compensate the absence of communication and acknowledgment packets. The prescribed performance of networked system is sustained in the presence of communication constraints. The probabilities of the data signal dropouts are expressed in terms of random Bernoulli processes. Then, simulation results are provided to demonstrate the validity of the proposed schemes and to compare them with others from literature. [ABSTRACT FROM AUTHOR]

We consider an online policy iteration-based reinforcement learning for a class of a nonlinear three-dimensional overhead crane with bounded uncertainties. Under the assumption that the rope length is fixed with small swing angles, a linearized model is derived. The system has four states; two actuated states: position x and y , and two un-actuated states, which are the rope angles θ x and θ y. The adaptive reinforcement learning controller is designed to handle the effects of measurement noises and outliers. We propose a model-free; hence it does not require precise knowledge of the system dynamics. When the state information is not available, a Kalman filter estimator is equipped with a dynamical saturation function to attenuate the effects of measurement noises and to remove outliers. A simulation study is established to illustrate the influence and robustness of the developed controller, and it can enhance the tracking trajectory under different scenarios to test the scheme. [ABSTRACT FROM AUTHOR]