DistributedControl - Distributed Control of Discrete Event...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Distributed Control of Discrete Event Systems Today many automated systems tend to be very large. For example a flexible manufacturing system may consists of 20 or 30 different workstations. The system may also be physically distributed across a large area such as the United States or perhaps the world. Therefore in order to control such a large and distributed system the need exists to use a collection of independent controllers that coordinates events by communicating with each other over the network. The following section describes how the simulation / control tool can be used to implement such a distributed controller. In the first section a logical method of decomposing the overall control task into sub task that are then implemented using individual controllers is presented.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The Control Hierarchy The control hierarchy was originally called Recursive Object Oriented Control Hierarchy (ROOCH) and was developed in collaboration with the Government Electronics Group at Motorola, Inc. and is published in Tirpak et al. (1992). The control hierarchy introduces a basic building block for modeling an automated system. The control object represents the most fundamental hierarchical element where control is to be implemented. It is assumed that each control object contains one or more subordinate processes P n (n=1,. ..,N), which are to execute tasks for the primary control object. If the subordinate controller is to be sent any item then the items enter the control object through its input port and will eventually exit through the output port. Items are assumed to be under the control of the control object from the moment they enter the input port, until they exit through the output port. Consequently, items residing in the output queue for the control object are controlled by the supervisor to the control object, because it is assumed that the control object no longer has any assigned tasks to be performed upon an item in the output queue. When the control object allocates a given item to a subordinate process P n , the physical control of that entity is relegated to the subordinate process to which it has been assigned. Note in Figure 1 that the Input Port and the Output Queue of each subordinate process belong to the control object, while the Input Queue and the Output Port for each subordinate process belong to that subordinate process. Therefore, a consistent chain of command for the control of a given entity has been defined, as the entity flows among the various unit processes within the control object until it eventually departs from the control object through its output port. To regulate the flows into the included input and output queues, inhibit flags have been specified. In general, the object to which the recipient queue belongs controls these flags. Thus, the Input Inhibit Flag is controlled by the subordinate process, while the Output Inhibit Flag is controlled by the supervisory control object.
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 8

DistributedControl - Distributed Control of Discrete Event...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online