José Yépez

Managing quality-of-control in network-based control systems by controller and message scheduling co-design

In network-based control systems (NCSs), plant sensor-controller-actuator nodes in closed-loop operation drive principal network traffic. The quality-of-control (QoC) in an NCS, i.e., the performance delivered by each closed-loop operation, depends not only on the controller design but also on the message scheduling strategy. In this paper, we show that the co-design of adaptive controllers and feedback scheduling policies allows for the optimization of the overall QoC. First, we discuss the limitations of standard discrete-time control models for controllers of control loops that are closed over communication networks. Afterwards, we describe an approach to adaptive controllers for NCS that: 1) overcomes some of the previous restrictions by online adapting the control decisions according to the dynamics of both the application and executing platform and 2) offers capabilities for dynamic management of QoC through message scheduling.

Control loop performance analysis over networked control systems

In this paper we discuss the suitability of standard discrete-time control models for the analysis and design of control loops implemented over networked control systems. In fact, depending on the specific control loop topology, different control approaches may apply. We distinguish between control approaches that offer synchronous and asynchronous actuation. Through simulation results we show that the performance of control loops closed over communication networks can be improved using approaches that support asynchronous actuation. This leads us to conclude that there is a need for more control approaches relying on the asynchronous actuation paradigm.

One-step finite horizon boundary with varying control gain for event-driven Networked Control Systems

Recent research has shown that event-driven control requires less samples than time-driven (periodic) control. Consequently, event-driven control applied to Networked Control Systems (NCS) is a good approach for alleviating controllers bandwidth demands and reducing overall network traffic. Following this trend, this paper presents an execution rule for event-driven networked controllers that at each job execution aims at postponing the next job execution while applying an appropriated controller gain considering each varying sampling interval that applies at run time. This method permits to dynamically lower the generated traffic for each networked control loop while ensuring the same or better control performance than the achieved by the periodic case. In addition, an implementation strategy capable of dealing with network induced time delays is also presented. Simulation results illustrate the operation and benefits of the presented approach.

Preliminary approach to Lyapunov sampling in CAN-based networked control systems

This paper presents a preliminary approach to networked control systems (NCS) that relies on an event-driven control method based on Lyapunov sampling. The goal is to study and develop approaches for NCS capable of offering controllers with low bandwidth demands. The paper starts by presenting the theoretical framework required for applying Lyapunov sampling to a set of closed-loop systems that share a serial bus line. The discussion permits identifying which requirements must be fulfilled in order to guarantee overall stability. In addition, the feasibility of the theoretical framework in terms of implementation requires to address problems caused by the implicit distributed architecture of the NCS. In particular, an efficient implementation approach is described for the CAN network. Simulation results illustrate the operation and benefits of the presented approach.

CICLIC: A Tool to Generate Feasible Cyclic Schedules

Summary A common way of implementing periodic real-time systems is by means of a cyclic scheduler. This method of scheduler is characterized by having a high degree of determinism, a high degree of predictability, a high degree of reliability, and implementation simplicity. Also, it is a very well-known method broadly used in industrial environments. The main problem this type of scheduler may have is the lack of flexibility, since any change in the group of tasks or in its temporary characteristics forces it to do repeat the execution plan. In solving these problems, it is necessary, and most of the time, essential, to have tools that may help system designers to create a cyclic scheduler automatically. In this paper, a tool having this function, and named CICLIC is described. The design mechanism is based on algorithms of exhaustive search that use heuristic rules in optimizing searching paths. To illustrate the use of CICLIC some examples are shown. Key Words : cyclic scheduler, real-time systems, simulation.

Hands-on course in networked control systems

This paper presents a hands-on course in networked control systems (NCS) to be integrated in the education of embedded control systems engineers. The course activities have a strong practical component and most of them are applied exercises to be implemented in a NCS setup. The paper describes the experimental setup and then proposes several activities that can be shaped into a course program according to the needs and diverse background of the targeted audience.