Engineering Real - Time Applications with P - NET

A recent trend in distributed computer-controlled systems (DCCS) is to interconnect the distributed elements by means of a multi-point broadcast network. As the network bus is shared between a number of network nodes, there is an access contention, which must be solved by the medium access control (MAC) protocol. Usually, DCCS impose real-time constraints. In essence, by real-time constraints we mean that traffic must be sent and received within a bounded interval, otherwise a timing fault is said to occur. This motivates the use of communication networks with MAC protocols that guarantee bounded access and response times to message requests. In this paper we analyse how the network access and queuing delays imposed by the P-NET's MAC mechanism can be evaluated in order to guarantee the real-time behaviour of the supported DCCS. 1. From Centralised to Distributed Computer Controlled Systems In the past decade manufacturing schemes have changed dramatically. In particular, the Computer Integrated Manufacturing (Groover, 1986) concept has been stressed as a means to achieve greater production competitiveness (Gutshke and Mertins, 1987). The driving forces behind the changes also resulted from the increased development and utilisation of new technologies that make massive use of microprocessor-based equipment. Integration implies that the different subsystems of the manufacturing environment interact and cooperate with each other. This means transfer, storage and processing of information in a widespread environment. In other words, integration requires efficient support for data communications. Nowadays, communication networks are available to virtually every aspect of the manufacturing environment, ranging from the production planning to the field level (Pimentel, 1990). However, the use of communication networks at the field level is a much more recent trend (Decotignie and Pleinevaux, 1993). Indeed, only more recently network interfaces become cost-effective for the interconnection of devices such as sensors and actuators, which, in the majority of the cases, are expected to be cheaper than the equipment typically interconnected at upper control levels of the manufacturing environment (e.g., workstations and numerically-controlled machines). The field level includes the process-relevant field devices, such as sensors and actuators. The process control level is hierarchically located above the field level, and directly influences it in the form of control signals. If the control is computer-based, the process control level uses the data received from the sensors to compute new commands, which are then transmitted to the actuators. A computer-controlled system can be decomposed into a set of three subsystems (Fig. 1): the controlled object; the computer system; the human operator (Kopetz, 1997). Collectively, the controlled object and the human operator can be referred to as the environment of the computer system. The interface between the human operator and the computer system is called the man-machine interface, and the interface between the controlled object and the computer system is called the instrumentation interface. The man-machine interface consists of input devices (e.g., keyboard) and output devices (e.g., display) that interface to the human operator. The role of the man-machine interface is to provide the computer system with, for example, process control set-points or device parameters for the sensors and actuators. It is also used to provide the operator with data for process control supervision. The instrumentation interface consists of the sensors and actuators that transform the physical signals in the controlled object into a digital form suitable to the computer system and vice-versa. * This work was partially supported by ISEP, FLAD, DEMEGI and FCT.