Electrical computers and digital processing systems: multicomput – Master/slave computer controlling
Reexamination Certificate
1999-11-17
2003-08-26
Coulter, Kenneth R. (Department: 2141)
Electrical computers and digital processing systems: multicomput
Master/slave computer controlling
C709S223000, C709S251000, C709S230000
Reexamination Certificate
active
06611860
ABSTRACT:
BACKGROUND OF THE INVENTION
1) Field of the Invention
The field of the invention pertains to methods and apparatus for implementing a control network and, more particularly, to a control network architecture and methods for connecting nodes in a control network.
2) Background
Automated control systems are commonly used in a number of manufacturing, transportation, and other applications, and are particularly useful to control machinery, sensors, electronics, and other system components. For example, manufacturing or vehicular systems may be outfitted with a variety of sensors and automated electrical and/or mechanical parts that require enablement or activation when needed to perform their predefined functions. Such systems commonly require that functions or procedures be carried out in a prescribed order or with a level of responsiveness that precludes sole reliance on manual control. Also, such systems may employ sensors or other components that require continuous or periodic monitoring and therefore lend themselves to automated control.
As the tasks performed by machinery have grown in number and complexity, a need has arisen for ways to exercise control over the various components of a system rapidly, efficiently and reliably. The sheer number of system components to be monitored, enabled, disabled, activated, deactivated, adjusted or otherwise controlled can lead to difficulties in designing and implementing a suitable control system. As the number of system components to be controlled is increased, not only is the operation of the control system made more complicated, but also the wiring and inter-connections of the control system are likewise more elaborate. In addition, greater reliance on automated control has resulted in larger potential consequences if the automated control system fails.
Traditionally, control systems in certain applications, such as transit vehicles and railcars, have relied upon relay-based control technology. In such systems, relays and switches are slaved to a logic circuit that serves to switch signal connections. This approach requires a large number of relays and a substantial amount of wiring throughout the vehicle. In some instances distributed processors or logic circuits may be used for subsystems such as the door, but these processors or logic circuits often take up significant space and can be costly to maintain.
A substantial improvement has recently been made in the field of control systems. An improved network control system recently developed uses a dual-bus architecture along with distributed controllers. In this improved network control system, a primary bus forms a high-speed, bi-directional communication link interconnecting a main data bus controller with distributed slave modules, one of which acts as a second data bus controller connected to a secondary, low-speed data bus. The slave modules are generally connected to various input/output ports. The second data bus controller can be connected to second-tier slave modules over the secondary, low-speed data bus. The main data bus controller, secondary data bus controller, first-tier slave modules, second-tier slave modules, input/output ports and other system components collectively form a hierarchical system wherein the main data bus controller supervises the first-tier slave modules, including the second data bus controller, the second data bus controller supervises the second-tier slave modules, and the first-tier slave modules and second-tier slave modules supervise their assigned input/output functions.
While the dual-bus control network as described above has many advantages, there are also ways in which it could be improved further. The dual-bus control network architecture as currently known in the art generally relies on a single top-level main data bus controller. If the main data bus controller fails, system performance will be adversely impacted. Also, the possibility of a short circuit occurring, particularly over a region of the bus, is a constant danger. In addition to disrupting communication signals among the components accessing the bus, a short circuit can be difficult to trace and cause substantial disruption of system service while maintenance personnel attempt to locate the short circuit. Furthermore, while the dual-bus network control architecture reduces wiring needed in a vehicle or other automated system, simplification of wiring connections would lead to greater ease of implementation and maintenance.
Accordingly, it would be advantageous to provide a network control system that has a means for recovering from a failure in a main data bus-controller or otherwise mitigating the effects such a failure. It would further be advantageous to provide a network control system that reduces the impact of a short circuit and enables rapid identification of the location of a short circuit by maintenance personnel. It would further be advantageous to provide a distributed network control system with simplified wiring and connections.
In addition, as hierarchical control networks grow in size and complexity, it becomes increasingly difficult for the master node or controller to communicate with nodes located remotely, separated often by a number of intervening, mid-level nodes or other intermediaries. Accordingly, it would be advantageous to provide a hierarchical control network which facilitates communication between the master node and lower level nodes.
SUMMARY OF THE INVENTION
According to one aspect of the invention, a control network is provided in which communication is facilitated between a master node and lower level nodes. In a preferred embodiment as described herein, a multiple-bus hierarchical control network includes a first-tier common bus and a plurality of lower-tier common buses. A first-tier master node controls a plurality of first-tier slave nodes using the first-tier common bus for communication. Each of the first-tier slave nodes may be connected to a separate second-tier common bus, and each operates as a respective second-tier master node for a plurality of second-tier slave nodes connected to the particular second-tier common bus associated with the first-tier slave/second-tier master node. Likewise, each of the second-tier slave nodes may be connected to a separate third-tier common bus, and each would then operate as a respective third-tier master node for a plurality of third-tier slave nodes connected to the particular third-tier common bus associated with the second-tier slave/third-tier master node. A supervisory network is connected to one or more of the lower-tier buses as well as to the first-tier common bus, so that activity on the lower-tier buses can be monitored by the first-tier master node.
In a preferred embodiment, the supervisory network comprises a first-tier supervisory master node connected to a common supervisory data bus. A plurality of supervisory slave nodes are connected to the supervisory data bus. Each of the supervisory slave nodes is also connected to one of the lower-tier buses, and each supervisory slave node monitors the communications occurring on the particular lower-tier bus. Information monitored by the supervisory slave nodes is communicated over the supervisory data bus to the first-tier supervisory master node either as needed, or on a regular basis, and can be relayed to the first-tier master node by the first-tier supervisory master node over the first-tier common bus.
If the control network is large and/or has many levels, then the supervisory network can be expanded by providing additional lower-tier supervisory levels, including lower-tier supervisory buses, configured with a control structure similar to the hierarchical control network absent the supervisory network. The resulting preferred network structure may be viewed as having a “matrix” architecture due to the intersection of control paths emanating downward from a single master node with the supervisory paths emanating across the control network from a single supervisory master node.
In a preferred embodiment, communication over a comm
Coulter Kenneth R.
I/O Controls Corporation
Irell & Manella LLP
LandOfFree
Control network with matrix architecture does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Control network with matrix architecture, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Control network with matrix architecture will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3126214