Patent Grant
7209812
The present invention relates to a system for controlling a model railroad.
Model railroads have traditionally been constructed with of a set of interconnected sections of train track, electric switches between different sections of the train track, and other electrically operated devices, such as train engines and draw bridges. Train engines receive their power to travel on the train track by electricity provided by a controller through the track itself. The speed and direction of the train engine is controlled by the level and polarity, respectively, of the electrical power supplied to the train track. The operator manually pushes buttons or pulls levers to cause the switches or other electrically operated devices to function, as desired. Such model railroad sets are suitable for a single operator, but unfortunately they lack the capability of adequately controlling multiple trains independently. In addition, such model railroad sets are not suitable for being controlled by multiple operators, especially if the operators are located at different locations distant from the model railroad, such as different cities.
A digital command control (DDC) system has been developed to provide additional controllability of individual train engines and other electrical devices. Each device the operator desires to control, such as a train engine, includes an individually addressable digital decoder. A digital command station (DCS) is electrically connected to the train track to provide a command in the form of a set of encoded digital bits to a particular device that includes a digital decoder. The digital command station is typically controlled by a personal computer. A suitable standard for the digital command control system is the NMRA DCC Standards, issued March 1997, and is incorporated herein by reference. While providing the ability to individually control different devices of the railroad set, the DCC system still fails to provide the capability for multiple operators to control the railroad devices, especially if the operators are remotely located from the railroad set and each other.
DigiToys Systems of Lawrenceville, Ga. has developed a software program for controlling a model railroad set from a remote location. The software includes an interface which allows the operator to select desired changes to devices of the railroad set that include a digital decoder, such as increasing the speed of a train or switching a switch. The software issues a command locally or through a network, such as the internet, to a digital command station at the railroad set which executes the command. The protocol used by the software is based on Cobra from Open Management Group where the software issues a command to a communication interface and awaits confirmation that the command was executed by the digital command station. When the software receives confirmation that the command executed, the software program sends the next command through the communication interface to the digital command station. In other words, the technique used by the software to control the model railroad is analogous to an inexpensive printer where commands are sequentially issued to the printer after the previous command has been executed. Unfortunately, it has been observed that the response of the model railroad to the operator appears slow, especially over a distributed network such as the internet. One technique to decrease the response time is to use high-speed network connections but unfortunately such connections are expensive.
What is desired, therefore, is a system for controlling a model railroad that effectively provides a high-speed connection without the additional expense associated therewith.
The foregoing and other objectives, features, and advantages of the invention will be more readily understood upon consideration of the following detailed description of the invention, taken in conjunction with the accompanying drawings.
The present invention overcomes the aforementioned drawbacks of the prior art, in a first aspect, by providing a system for operating a digitally controlled model railroad that includes transmitting a first command from a first client program to a resident external controlling interface through a first communications transport. A second command is transmitted from a second client program to the resident external controlling interface through a second communications transport. The first command and the second command are received by the resident external controlling interface which queues the first and second commands. The resident external controlling interface sends third and fourth commands representative of the first and second commands, respectively, to a digital command station for execution on the digitally controlled model railroad.
Incorporating a communications transport between the multiple client program and the resident external controlling interface permits multiple operators of the model railroad at locations distant from the physical model railroad and each other. In the environment of a model railroad club where the members want to simultaneously control devices of the same model railroad layout, which preferably includes multiple trains operating thereon, the operators each provide commands to the resistant external controlling interface, and hence the model railroad. In addition by queuing by commands at a single resident external controlling interface permits controlled execution of the commands by the digitally controlled model railroad, would may otherwise conflict with one another.
In another aspect of the present invention the first command is selectively processed and sent to one of a plurality of digital command stations for execution on the digitally controlled model railroad based upon information contained therein. Preferably, the second command is also selectively processed and sent to one of the plurality of digital command stations for execution on the digitally controlled model railroad based upon information contained therein. The resident external controlling interface also preferably includes a command queue to maintain the order of the commands.
The command queue also allows the sharing of multiple devices, multiple clients to communicate with the same device (locally or remote) in a controlled manner, and multiple clients to communicate with different devices. In other words, the command queue permits the proper execution in the cases of: (1) one client to many devices, (2) many clients to one device, and (3) many clients to many devices.
In yet another aspect of the present invention the first command is transmitted from a first client program to a first processor through a first communications transport. The first command is received at the first processor. The first processor provides an acknowledgement to the first client program through the first communications transport indicating that the first command has properly executed prior to execution of commands related to the first command by the digitally controlled model railroad. The communications transport is preferably a COM or DCOM interface.
The model railroad application involves the use of extremely slow real-time interfaces between the digital command stations and the devices of the model railroad. In order to increase the apparent speed of execution to the client, other than using high-speed communication interfaces, the resident external controller interface receives the command and provides an acknowledgement to the client program in a timely manner before the execution of the command by the digital command stations. Accordingly, the execution of commands provided by the resident external controlling interface to the digital command stations occur in a synchronous manner, such as a first-in-first-out manner. The COM and DCOM communications transport between the client program and the resident external controlling interface is operated in an asynchronous manner, namely providing an acknowledgement thereby releasing the communications is transport to accept further communications prior to the actual execution of the command. The combination of the synchronous and the asynchronous data communication for the commands provides the benefit that the operator considers the commands to occur nearly instantaneously while permitting the resident external controlling interface to verify that the command is proper and cause the commands to execute in a controlled manner by the digital command stations, all without additional high-speed communication networks. Moreover, for traditional distributed software execution there is no motivation to provide an acknowledgment prior to the execution of the command because the command executes quickly and most commands are sequential in nature. In other words, the execution of the next command is dependent upon proper execution of the prior command so there would be no motivation to provide an acknowledgment prior to its actual execution.
Referring to
The communications transport 12 provides an interface between the client program 14 and the resident external controlling interface 16. The communications transport 12 may be any suitable communications medium for the transmission of data, such as the internet, local area network, satellite links, or multiple processes operating on a single computer. The preferred interface to the communications transport 12 is a COM or DCOM interface, as developed for the Windows operating system available from Microsoft Corporation. The communications transport 12 also determines if the resident external controlling interface 16 is system resident or remotely located on an external system. The communications transport 12 may also use private or public communications protocol as a medium for communications. The client program 14 provides commands and the resident external controlling interface 16 responds to the communications transport 12 to exchange information. A description of COM (common object model) and DCOM (distributed common object model) is provided by Chappel in a book entitled Understanding ActiveX and OLE, Microsoft Press, and is incorporated by reference herein.
Incorporating a communications transport 12 between the client program(s) 14 and the resident external controlling interface 16 permits multiple operators of the model railroad at locations distant from the physical model railroad and each other. In the environment of a model railroad club where the members want to simultaneously control devices of the same model railroad layout, which preferably includes multiple trains operating thereon, the operators each provide commands to the resistant external controlling interface, and hence the model railroad.
The manner in which commands are executed for the model railroad under COM and DCOM may be as follows. The client program 14 makes requests in a synchronous manner using COM/DCOM to the resident external interface controller 16. The synchronous manner of the request is the technique used by COM and DCOM to execute commands. The communications transport 12 packages the command for the transport mechanism to the resident external controlling interface 16. The resident external controlling interface 16 then passes the command to the digital command stations 18 which in turn executes the command. After the digital command station 18 executes the command an acknowledgement is passed back to the resident external controlling interface 16 which in turn passes an acknowledgement to the client program 14. Upon receipt of the acknowledgement by the client program 14, the communications transport 12 is again available to accept another command. The train control system 10, without more, permits execution of commands by the digital command stations 18 from multiple operators, but like the DigiToys Systems' software the execution of commands is slow.
The present inventor came to the realization that unlike traditional distributed systems where the commands passed through a communications transport are executed nearly instantaneously by the server and then an acknowledgement is returned to the client, the model railroad application involves the use of extremely slow real-time interfaces between the digital command stations and the devices of the model railroad. The present inventor came to the further realization that in order to increase the apparent speed of execution to the client, other than using high-speed communication interfaces, the resident external controller interface 16 should receive the command and provide an acknowledgement to the client program 12 in a timely manner before the execution of the command by the digital command stations 18. Accordingly, the execution of commands provided by the resident external controlling interface 16 to the digital command stations 18 occur in a synchronous manner, such as a first-in-first-out manner. The COM and DCOM communications transport 12 between the client program 14 and the resident external controlling interface 16 is operated in an asynchronous manner, namely providing an acknowledgement thereby releasing the communications transport 12 to accept further communications prior to the actual execution of the command. The combination of the synchronous and the asynchronous data communication for the commands provides the benefit that the operator considers the commands to occur nearly instantaneously while permitting the resident external controlling interface 16 to verify that the command is proper and cause the commands to execute in a controlled manner by the digital command stations 18, all without additional high-speed communication networks. Moreover, for traditional distributed software execution there is no motivation to provide an acknowledgment prior to the execution of the command because the command executes quickly and most commands are sequential in nature. In other words, the execution of the next command is dependent upon proper execution of the prior command so there would be no motivation to provide an acknowledgment prior to its actual execution. It is to be understood that other devices, such as digital devices, may be controlled in a manner as described for model railroads.
Referring to
The asynchronous command processor 100 also verifies, using the configuration information in the local database storage 102, that the command received is a potentially valid operation. If the command is invalid, the asynchronous command processor 100 provides such information to the asynchronous response processor 106, which in turn returns an error indication to the client program 14.
The asynchronous command processor 100 may determine that the necessary information is not contained in the local database storage 102 to provide a response to the client program 14 of the device state or that the command is a valid action. Actions may include, for example, an increase in the train's speed, or turning on/off of a device. In either case, the valid unknown state or action command is packaged and forwarded to the command queue 104. The packaging of the command may also include additional information from the local database storage 102 to complete the client program 14 request, if necessary. Together with packaging the command for the command queue 104, the asynchronous command processor 100 provides a command to the asynchronous request processor 106 to provide a response to the client program 14 indicating that the event has occurred, even though such an event has yet to occur on the physical railroad layout.
As such, it can be observed that whether or not the command is valid, whether or not the information requested by the command is available to the asynchronous command processor 100, and whether or not the command has executed, the combination of the asynchronous command processor 100 and the asynchronous response processor 106 both verifies the validity of the command and provides a response to the client program 14 thereby freeing up the communications transport 12 for additional commands. Without the asynchronous nature of the resident external controlling interface 16, the response to the client program 14 would be, in many circumstances, delayed thereby resulting in frustration to the operator that the model railroad is performing in a slow and painstaking manner. In this manner, the railroad operation using the asynchronous interface appears to the operator as nearly instantaneously responsive.
Each command in the command queue 104 is fetched by a synchronous command processor 110 and processed. The synchronous command processor 110 queries a controller database storage 112 for additional information, as necessary, and determines if the command has already been executed based on the state of the devices in the controller database storage 112. In the event that the command has already been executed, as indicated by the controller database storage 112, then the synchronous command processor 110 passes information to the command queue 104 that the command has been executed or the state of the device. The asynchronous response processor 106 fetches the information from the command cue 104 and provides a suitable response to the client program 14, if necessary, and updates the local database storage 102 to reflect the updated status of the railroad layout devices.
If the command fetched by the synchronous command processor 110 from the command queue 104 requires execution by external devices, such as the train engine, then the command is posted to one of several external device control logic 114 blocks. The external device control logic 114 processes the command from the synchronous command processor 110 and issues appropriate control commands to the interface of the particular external device 116 to execute the command on the device and ensure that an appropriate response was received in response. The external device is preferably a digital command control device that transmits digital commands to decoders using the train track. There are several different manufacturers of digital command stations, each of which has a different set of input commands, so each external device is designed for a particular digital command station. In this manner, the system is compatible with different digital command stations. The digital command stations 18 of the external devices 116 provide a response to the external device control logic 114 which is checked for validity and identified as to which prior command it corresponds to so that the controller database storage 112 may be updated properly. The process of transmitting commands to and receiving responses from the external devices 116 is slow.
The synchronous command processor 110 is notified of the results from the external control logic 114 and, if appropriate, forwards the results to the command queue 104. The asynchronous response processor 100 clears the results from the command queue 104 and updates the local database storage 102 and sends an asynchronous response to the client program 14, if needed. The response updates the client program 14 of the actual state of the railroad track devices, if changed, and provides an error message to the client program 14 if the devices actual state was previously improperly reported or a command did not execute properly.
The use of two separate database storages, each of which is substantially a mirror image of the other, provides a performance enhancement by a fast acknowledgement to the client program 14 using the local database storage 102 and thereby freeing up the communications transport 12 for additional commands. In addition, the number of commands forwarded to the external device control logic 114 and the external devices 116, which are relatively slow to respond, is minimized by maintaining information concerning the state and configuration of the model railroad. Also, the use of two separate database tables 102 and 112 allows more efficient multi-threading on multi-processor computers.
In order to achieve the separation of the asynchronous and synchronous portions of the system the command queue 104 is implemented as a named pipe, as developed by Microsoft for Windows. The queue 104 allows both portions to be separate from each other, where each considers the other to be the destination device. In addition, the command queue maintains the order of operation which is important to proper operation of the system.
The use of a single command queue 104 allows multiple instantrations of the asynchronous functionality, with one for each different client. The single command queue 104 also allows the sharing of multiple devices, multiple clients to communicate with the same device (locally or remote) in a controlled manner, and multiple clients to communicate with different devices. In other words, the command queue 104 permits the proper execution in the cases of: (1) one client to many devices, (2) many clients to one device, and (3) many clients to many devices.
The present inventor came to the realization that the digital command stations provided by the different vendors have at least three different techniques for communicating with the digital decoders of the model railroad set. The first technique, generally referred to as a transaction (one or more operations), is a synchronous communication where a command is transmitted, executed, and a response is received therefrom prior to the transmission of the next sequentially received command. The DCS may execute multiple commands in this transaction. The second technique is a cache with out of order execution where a command is executed and a response received therefrom prior to the execution of the next command, but the order of execution is not necessarily the same as the order that the commands were provided to the command station. The third technique is a local-area-network model where the commands are transmitted and received simultaneously. In the LAN model there is no requirement to wait until a response is received for a particular command prior to sending the next command. Accordingly, the LAN model may result in many commands being transmitted by the command station that have yet to be executed. In addition, some digital command stations use two or more of these techniques.
With all these different techniques used to communicate with the model railroad set and the system 10 providing an interface for each different type of command station, there exists a need for the capability of matching up the responses from each of the different types of command stations with the particular command issued for record keeping purposes. Without matching up the responses from the command stations, the databases can not be updated properly.
Validation functionality is included within the external device control logic 114 to accommodate all of the different types of command stations. Referring to
Building your own visual interface to a model railroad Copyright 1992–1998 KAM Industries. Computer Dispatcher, Engine Commander, The Conductor, Train Server, and Train Tools are Trademarks of KAM Industries, all Rights Reserved. Questions concerning the product can be EMAILED to: traintools@kam.rain.com You can also mail questions to: KAM Industries 2373 NW 185th Avenue Suite 416 Hillsboro, Oreg. 97124 FAX-(503) 291-1221
This document is divided into two sections, the Tutorial, and the IDL Command Reference. The tutorial shows the complete code for a simple Visual BASIC program that controls all the major functions of a locomotive. This program makes use of many of the commands described in the reference section. The IDL Command Reference describes each command in detail.
A. Visual BASIC Throttle Example Application
The following application is created using the Visual BASIC source code in the next section. It controls all major locomotive functions such as speed, direction, and auxiliary functions.
A. Visual BASIC Throttle Example Source Code
A. Introduction
This document describes the IDL interface to the KAM Industries Engine Commander Train Server. The Train Server DCOM server may reside locally or on a network node This server handles all the background details of controlling your railroad. You write simple, front end programs in a variety of languages such as BASIC, Java, or C++ to provide the visual interface to the user while the server handles the details of communicating with the command station, etc.
A. Data Types
Data is passed to and from the IDL interface using a several primitive data types. Arrays of these simple types are also used. The exact type passed to and from your program depends on the programming language your are using.
The following primitive data types are used:
A long /DecoderObject/D value is returned by the KamDecoderPutAdd call if the decoder is successfully registered with the server. This unique opaque ID should be used for all subsequent calls to reference this decoder.
A. Commands to Access the Server Configuration Variable Database
This section describes the commands that access the server configuration variables (CV) database. These CVs are stored in the decoder and control many of its characteristics such as its address. For efficiency, a copy of each CV value is also stored in the server database. Commands such as KamCVGetValue and KamCVPutValue communicate only with the server, not the actual decoder. You then use the programming commands in the next section to transfer CVs to and from the decoder.
KamCVGetValue takes the decoaer object ID and con figuration variable (CV) number as parameters. It sets the memory pointed to by pCVValue to the value of the server copy of the configuration variable.
KamCVPutValue takes the decoder object ID, configuration variable (CV) number, and a new CV value as parameters. It sets the server copy of the specified decoder CV to iCVValue.
KamCVGetEnable takes the decoder object ID, configuration variable (CV) number, and a pointer to store the enable flag as parameters. It sets the location pointed to by pEnable.
KamCVPutEnable takes the decoder object ID, configuration variable (CV) number, and a new enable state as parameters. It sets the server copy of the CV bit mask to iEnable.
KamCVGetName takes a configuration variable (CV) number as a parameter. It sets the memory pointed to by pbsCVNameString to the name of the CV as defined in NMRA Recommended Practice RP 9.2.2.
KamCVGetMinRegister takes a decoder object ID as a parameter. It sets the memory pointed to by pMinRegister to the minimum possible CV register number for the specified decoder.
KamCVGetMaxRegister takes a decoder object ID as a parameter. It sets the memory pointed to by pMaxRegister to the maximum possible CV register number for the specified decoder.
A. Commands to Program Configuration Variables
This section describes the commands read and write decoder configuration variables (CVs). You should initially transfer a copy of the decoder CVs to the server using the KamProgramReadDecoderToDataBase command. You can then read and modify this server copy of the CVs. Finally, you can program one or more CVs into the decoder. using the KamProgramCV or KamProgramDecoderFromDataBase command. Not that you must first enter programming mode by issuing the KamProgram command before any programming can be done.
KamProgram take the decoder object ID, logical programming port ID, and programming mode as parameters. It changes the command station mode from normal operation (PROGRAM_MODE_NONE) to the specified programming mode. Once in programming modes, any number of programming commands may be called. When done, you must call KamProgram with a parameter of PROGRAM_MODE_NONE to return to normal operation.
KamProgramGetMode take the decoder object ID, logical programming port ID, and pointer to a place to store the programming mode as parameters. It sets the memory pointed to by piProgMode to the present programming mode.
KamProgramGetStatus take the decoder object ID and pointer to a place to store the OR'd decoder programming status as parameters. It sets the memory pointed to by piProgMode to the present programming mode.
KamProgramCV takes the decoder object ID, configuration variable (CV) number as parameters. It reads the specified CV variable value to the server database.
KamProgramCV takes the decoder object ID, configuration variable (CV) number, and a new CV value as parameters. It programs (writes) a single decoder CV using the specified value as source data.
KamProgramReadDecoderToDataBase takes the decoder object ID as a parameter. It reads all enabled CV values from the decoder and stores them in the server database.
KamProgramDecoderFromDataBase takes the decoder object ID as a parameter. It programs (writes) all enabled decoder CV values using the server copy of the CVs as source data.
A. Commands to Control all Decoder Types
This section describes the commands that all decoder types. These commands do things such getting the maximum address a given type of decoder supports, adding decoders to the database, etc.
KamDecoderGetMaxModels takes no parameters. It sets the memory pointed to by piMaxModels to the maximum decoder type ID.
KamPortGetModelName takes a decoder type ID and a pointer to a string as parameters. It sets the memory pointed to by pbsModelName to a BSTR containing the decoder name.
KamDecoderSetModelToObj takes a decoder ID and decoder object ID as parameters. It sets the decoder model type of the decoder at address IDecoderObjectID to the type specified by iModel.
KamDecoderGetMaxAddress takes a decoder type ID and a pointer to store the maximum address as parameters. It sets the memory pointed to by piMaxAddress to the maximum address supported by the specified decoder.
KamDecoderChangeOldNewAddr takes an old decoder object ID and a new decoder address as parameters. It moves the specified locomotive or accessory decoder to iNewAddr and sets the memory pointed to by piNewObjID to the new object ID. The old object ID is now invalid and should no longer be used.
KamDecoderMovePort takes a decoder object ID and logical port ID as parameters. It moves the decoder specified by IDecoderObjectID to the controller specified by iLogicalPortID.
KamDecoderMovePort takes a decoder object ID and pointer to a logical port ID as parameters. It sets the memory pointed to by piLogicalPortID to the logical port ID associated with lDecoderObjectID.
IDS_ERR_ADDRESSEXIST returned if call succeeded but the address exists. KamDecoderCheckAddrinUse takes a decoder address, logical port, and decoder class as parameters. It returns zero if the address is not in use. It will return IDS_ERR_ADDRESSEXIST if the call succeeds but the address already exists. It will return the appropriate non zero error number if the calls fails.
KamDecoderGetModelFromObj takes a decoder object ID and pointer to a decoder type ID as parameters. It sets the memory pointed to by piModel to the decoder type ID associated with iDCCAddr.
KamDecoderGetModelFacility takes a decoder object ID and pointer to a decoder facility, mask as parameters. It sets the memory pointed to by pdwFacility to the decoder facility mask associated with iDCCAddr.
KamuecoderGetObjCount takes a decoder class and a pointer to an address count as parameters. It sets the memory pointed to by piObjCount to the count of active decoders of the type given by iDecoderClass.
KamDecoderGetObjCount takes a decoder index, decoder class, and a pointer to an object ID as parameters It sets the memory pointed to by plDecoderObjectID to the selected object ID.
KamDecoderPutAdd takes a decoder object ID, command logical port, programming logical port, clear flag, decoder model ID, and a pointer to a decoder object ID as parameters. It creates a new locomotive object in the locomotive database and sets the memory pointed to by plDecoderObjectID to the decoder object ID used by the server as a key.
KamDecoderPutDel takes a decoder object ID and clear flag as parameters. It deletes the locomotive object specified by IDecoderObjectID from the locomotive database.
KamDecoderGetMfgName takes a decoder object ID and pointer to a manufacturer name string as parameters. It sets the memory pointed to by pbsMfgName to the name of the decoder manufacturer.
KamDecoderGetPowerMode takes a decoder object ID and a pointer to the power mode string as parameters. It sets the memory pointed to by pbsPowerMode to the decoder power mode.
KamDecoderGetMaxSpeed takes a decoder object ID and a pointer to the maximum supported speed step as parameters. It sets the memory pointed to by piSpeedStep to the maximum speed step supported by the decoder.
A. Commands to Control Locomotive Decoders
This section describes the commands that control locomotive decoders. These commands control things such as locomotive speed and direction. For efficiency, a copy of all the engine variables such speed communicate only with the server, not the actual decoder. You should first make any changes to the server copy of the engine variables. You can send all changes to the engine using the KamCmdCommand command.
KamEngGetSpeed takes the decoder object ID and pointers to locations to store the locomotive speed and direction as parameters. It sets the memory pointed to by lpSpeed to the locomotive speed and the memory pointed to by lpDirection to the locomotive direction.
KamEngPutSpeed takes the decoder object ID, new locomotive speed, and new locomotive direction as parameters. It sets the locomotive database speed to iSpeed and the locomotive database direction to iDirection. Note: This command only changes the locomotive database. The data is not sent to the decoder until execution of the KamCmdCommand command. Speed is set to the maximum possible for the decoder if iSpeed exceeds the decoders range.
KamEngGetSpeedSteps takes the decoder object ID and a pointer to a location to store the number of speed steps as a parameter. It sets the memory pointed to by lpSpeedSteps to the number of speed steps.
KamEngPutSpeedSteps takes the decoder object ID and a new number of speed steps as a parameter. It sets the number of speed steps in the locomotive database to iSpeedSteps. Note: This command only changes the locomotive database. The data is not sent to the decoder until execution of the KamCmdCommand command. KamDecoderGetMaxSpeed returns the maximum possible speed for the decoder. An error is generated if an attempt is made to set the speed steps beyond this value.
KamEngGetFunction takes the decoder object ID, a function ID, and a pointer to the location to store the specified function state as parameters. It sets the memory pointed to by lpFunction to the specified function state.
KamEngPutFunction takes the decoder object ID, a function ID, and a new function state as parameters. It sets the specified locomotive database function state to iFunction. Note: This command only changes the locomotive database. The data is not sent to the decoder until execution of the KamCmdCommand command.
KamEngGetFunctionMax takes a decoder object ID and a pointer to the maximum function ID as parameters. It sets the memory pointed to by piMaxFunction to the maximum possible function number for the specified decoder.
KamEngGetName takes a decoder object ID and a pointer to the locomotive name as parameters. It sets the memory pointed to by pbsEngName to the name of the locomotive.
KamEngPutName taxes a decoder object ID and a BSTR as parameters. It sets the symbolic locomotive name to bsEngName.
KamEngGetFuncntionName takes a decoder object ID, function ID, and a pointer to the function name as parameters. It sets the memory pointed to by pbsFcnNameString to the symbolic name of the specified function.
KamEngPutFunctionName takes a decoder object ID, function ID, and a BSTR as parameters. It sets the specified symbolic function name to bsFcnNameString.
KamEngGetConsistMax takes the decoder object ID and a pointer to a location to store the maximum consist as parameters. It sets the location pointed to by piMaxConsist to the maximum number of locomotives that can but placed in a command station controlled consist. Note that this command is designed for command station consisting. CV consisting is handled using the CV commands.
KamEngPutConsistParent takes the parent object ID and an alias addres as parameter. It makes the decoder specified by lDCCParentObjID the consist parent referred to by iDCCAliasAddr. Note that this command is designed for command station consisting. CV consisting is handled using the CV commands. If a new parent is defined for a consist; the old parent becomes a child in the consist. To delete a parent in a consist without deleting the consist, you must add a new parent then delete the old parent using KamEngPutConsistRemoveObj.
KamEngPutConsistChild takes the decoder parent object ID and decoder object ID as parameters. It assigns the decoder specified by lDCCObjID to the consist identified by lDCCParentObjID. Note that this command is designed for command station consisting. CV consisting is handled using the CV commands. Note: This command is invalid if the parent has not been set previously using KamEngPutConsistParent.
KamEngPutConsistRemoveObj takes the decoder object ID as a parameter. It removes the decoder specified by lDecoderObjectID from the consist. Note that this command is designed for command station consisting. CV consisting is handled using the CV commands. Note: If the parent is removed, all children are removed also.
A. Commands to Control Accessory Decoders
This section describes the commands that control accessory decoders. These commands control things such as accessory decoder activation state. For efficiency, a copy of all the engine variables such speed is stored in the server. Commands such as KamAccGetFunction communicate only with the server, not the actual decoder. You should first make any changes to the server copy of the engine variables. You can send all changes to the engine using the KamCmdCommand command.
KamAccGetFunction takes the decoder object ID, a function ID, and a pointer to the location to store the specified function state as parameters. It sets the memory pointed to by lpFunction to the specified function state.
KanAccGetFunctionAll takes the decoder object ID and a pointer to a bit mask as parameters. It sets each bit in the memory pointed to by piValue to the corresponding function state.
KamAccPutFunction takes the decoder object ID, a function ID, and a new function state as parameters. It sets the specified accessory database function state to iFunction. Note: This command only changes the accessory database. The data is not sent to the decoder until execution of the KamCmdCommand command.
KamAccPutFunctionAll takes the decoder object ID and a bit mask as parameters. It sets all decoder function enable states to match the state bits in iValue. The possible enable states are TRUE and FALSE. The data is not sent to the decoder until execution of the KamCmdCommand command.
KamAccGetFunctionMax takes a decoder object ID and pointer to the maximum function number as parameters. It sets the memory pointed to by piMaxFunction to the maximum possible function number for the specified decoder.
KamAccGetName takes a decoder object ID and a pointer to a string as parameters. It sets the memory pointed to by pbsAccNameString to the name of the accessory.
KamAccPutName takes a decoder object ID and a BSTR as parameters. It sets the symbolic accessory name to bsAccName.
KamAccGetFuncntionName takes a decoder object ID, function ID, and a pointer to a string as parameters. It sets the memory pointed to by pbsAcnNameString to the symbolic name of the specified function.
KamAccPutFunctionName takes a decoder object ID, function ID, and a BSTR as parameters. It sets the specified symbolic function name to bsFcnNameString.
KamAccRegFeedback takes a decoder object ID, node name string, and function ID, as parameters. It registers interest in the function given by iFunctionID by the method given by the node name string bsAccNode. bsAccNode identifies the server application and method to call if the function changes state. Its format is “\\{Server}\{App}.{Method}” where {Server} is the server name, {App} is the application name, and {Method} is the method name.
KamAccRegFeedbackAll takes a decoder object ID and node name string as parameters. It registers interest in all functions by the method given by the node name string bsAccNode. bsAccNode identifies the server application and method to call if the function changes state. Its format is “\\{Server}\{App}.{Method}” where {Server} is the server name, {App} is the application name, and {Method} is the method name.
KamAccDelFeedback takes a decoder object ID, node name string, and function ID, as parameters. It deletes interest in the function given by iFunctionID by the method given by the node name string bsAccNode. bsAccNode identifies the server application and method to call if the function changes state. Its format is “\\{Server}\{App}.{Method}” where {Server} is the server name, {App} is the application name, and {Method} is the method name.
KamAccDelFeedbackAll takes a decoder object ID and node name string as parameters. It deletes interest in all functions by the method given by the node name string bsAccNode. bsAccNode identifies the server application and method to call if the function changes state. Its format is “\\{Server}\{App}.{Method}” where {Server} is the server name, {App} is the application name, and {Method} is the method name.
A. Commands to Control the Command Station
This section describes the commands that control the command station. These commands do things such as controlling command station power. The steps to control a given command station vary depending on the type of command station.
KamOprPutTurnOnStation takes a logical port ID as a parameter. It performs the steps necessary to turn on the command station. This command performs a combination of other commands such as KamOprPutStartStation, KamOprPutClearStation, and KamOprPutPowerOn.
KamOprPutStartStation takes a logical port ID as a parameter. It performs the steps necessary to start the command station.
KamOprPutClearStation takes a logical port ID as a parameter. It performs the steps necessary to clear the command station queue.
KamOprPutStopStation takes a logical port ID as a parameter. It performs the steps necessary to stop the command station.
KamOprPutPowerOn takes a logical port ID as a parameter. It performs the steps necessary to apply power to the track.
KamOprPutPowerOff takes a logical port ID as a parameter. It performs the steps necessary to remove power from the track.
KamOprPutHardReset takes a logical port ID as a parameter. It performs the steps necessary to perform a hard reset of the command station.
KamOprPutEmergencyStop takes a logical port ID as a parameter. It performs the steps necessary to broadcast an emergency stop command to all decoders.
KamOprGetStationStatus takes a logical port ID and a pointer to a string as parameters. It set the memory pointed to by pbsCmdStat to the command station status. The exact format of the status BSTR is vendor dependent.
A. Commands to Configure the Command Station Communication Port
This section describes the commands that configure the command station communication port. These commands do things such as setting BAUD rate. Several of the commands in this section use the numeric controller ID (iControllerID) to identify a specific type of command station controller. The following table shows the mapping between the controller ID (iControllerID) and controller name (bsControllerName) for a given type of command station controller.
KamPortPutConfig takes a logical port ID, configuration index, configuration value, and key as parameters. It sets the port parameter specified by iIndex to the value specified by iValue. For the DEBUG iIndex value, the debug file path is C:\Temp\Debug{PORT}.txt where {PORT} is the physical comm port ID.
KamPortGetConfig takes a logical port ID, configuration index, and a pointer to a configuration value as parameters. It sets the memory pointed to by piValue to the specified configuration value.
KamPortGetName takes a physical port ID number and a pointer to a port name string as parameters. It sets the memory pointed to by pbsPortName to the physical port name such as “COMM1.”
KamPortPutMapController takes a logical port ID, a command station type ID, and a physical communications port ID as parameters. It maps iLogicalPortID to iCommPortID for the type of command station specified by iControllerID.
KamPortGetMaxLogPorts takes a pointer to a logical port ID as a parameter. It sets the memory pointed to by piMaxLogicalPorts to the maximum logical port ID.
KamPortGetMaxPhysical takes a pointer to the number of physical ports, the number of serial ports, and the number of parallel ports as parameters. It sets the memory pointed to by the parameters to the associated values
A. Commands that Control Command flow to the Command Station
This section describes the commands that control the command flow to the command station. These commands do things such as connecting and disconnecting from the command station.
KamCmdConnect takes a logical port ID as a parameter. It connects the server to the specified command station.
KamCmdDisConnect takes a logical port ID as a parameter. It disconnects the server to the specified command station.
KamCmdCommand takes the decoder object ID as a parameter. It sends all state changes from the server database to the specified locomotive or accessory decoder.
A. Cab Control Commands
This section describes commands that control the cabs attached to a command station.
KamCabGetMessage takes a cab address and a pointer to a message string as parameters. It sets the memory pointed to by pbsMsg to the present cab message.
KamCabPutMessage takes a cab address and a BSTR as parameters. It sets the cab message to bsMsg.
KamCabGetCabAddr takes a decoder object ID and a pointer to a cab address as parameters. It set the memory pointed to by piCabAddress to the address of the cab attached to the specified decoder.
KamCabPutAddrToCab takes a decoder object ID and cab address as parameters. It attaches the decoder specified by iDCCAddr to the cab specified by iCabAddress.
A. Miscellaneous Commands
This section describes miscellaneous commands that do not fit into the other categories.
KamMiscGetErrorMsg takes an error flag as a parameter. It returns a BSTR containing the descriptive error message associated with the specified error flag.
KamMiscGetClockTime takes the port ID, the time mode, and pointers to locations to store the day, hours, minutes, and fast clock ratio as parameters. It sets the memory pointed to by piDay to the fast clock day, sets pointed to by piHours to the fast clock hours, sets the memory pointed to by piMinutes to the fast clock minutes, and the memory pointed to by piRatio to the fast clock ratio. The servers local time will be returned if the command station does not support a fast clock.
KamMiscPutClockTime takes the fast clock logical port, the fast clock day, the fast clock hours, the fast clock minutes, and the fast clock ratio as parameters. It sets the fast clock using specified parameters.
KamMiscGetInterfaceVersion takes a pointer to an interface version string as a parameter. It sets the memory pointed to by pbsInterfaceVersion to the interface version string. The version string may contain multiple lines depending on the number of interfaces supported.
KamMiscSaveData takes no parameters. It saves all server data to permanent storage. This command is run automatically whenever the server stops running. Demo versions of the program cannot save data and this command will return an error in that case.
KamMiscGetControllerName takes a command station type ID and a pointer to a type name string as parameters. It sets the memory pointed to by pbsName to the command station type name.
KamMiscGetControllerName takes a logical port ID and a pointer to a command station type name as parameters. It sets the memory pointed to by pbsName to the command station type name for that logical port.
KamMiscGetCommandStationValue takes the controller ID, logical port, value array index, and a pointer to the location to store the selected value. It sets the memory pointed to by piValue to the specified command station miscellaneous data value.
KamMiscSetCommandStationValue takes the controller ID, logical port, value array index, and new miscellaneous data value. It sets the specified command station data to the value given by piValue.
KamMiscGetCommandStationIndex takes the controller ID, logical port, and a pointer to the location to store the maximum index. It sets the memory pointed to by piIndex to the specified command station maximum miscellaneous data index.
KamMiscMaxControllerID takes a pointer to the maximum controller ID as a parameter. It sets the memory pointed to by piMaxControllerID to the maximum controller type ID.
KamMiscGetControllerFacility takes the controller ID and a pointer to the location to store the selected controller facility mask. It sets the memory pointed to by pdwFacility to the specified command station facility mask.
The digital command stations 18 program the digital devices, such as a locomotive and switches, of the railroad layout. For example, a locomotive may include several different registers that control the horn, how the light blinks, speed curves for operation, etc. In many such locomotives there are 106 or more programable values. Unfortunately, it may take 1–10 seconds per byte wide word if a valid register or control variable (generally referred to collectively as registers) and two to four minutes to error out if an invalid register to program such a locomotive or device, either of which may contain a decoder. With a large number of byte wide words in a locomotive its takes considerable time to fully program the locomotive. Further, with a railroad layout including many such locomotives and other programmable devices, it takes a substantial amount of time to completely program all the devices of the model railroad layout. During the programming of the railroad layout, the operator is sitting there not enjoying the operation of the railroad layout, is frustrated, loses operating enjoyment, and will not desire to use digital programmable devices. In addition, to reprogram the railroad layout the operator must reprogram all of the devices of the entire railroad layout which takes substantial time. Similarly, to determine the state of all the devices of the railroad layout the operator must read the registers of each device likewise taking substantial time. Moreover, to reprogram merely a few bytes of a particular device requires the operator to previously know the state of the registers of the device which is obtainable by reading the registers of the device taking substantial time, thereby still frustrating the operator.
The present inventor came to the realization that for the operation of a model railroad the anticipated state of the individual devices of the railroad, as programmed, should be maintained during the use of the model railroad and between different uses of the model railroad. By maintaining data representative of the current state of the device registers of the model railroad determinations may be made to efficiently program the devices. When the user designates a command to be executed by one or more of the digital command stations 18, the software may determine which commands need to be sent to one or more of the digital command stations 18 of the model railroad. By only updating those registers of particular devices that are necessary to implement the commands of a particular user, the time necessary to program the railroad layout is substantially reduced. For example, if the command would duplicate the current state of the device then no command needs to be forwarded to the digital command stations 18. This prevents redundantly programming the devices of the model railroad, thereby freeing up the operation of the model railroad for other activities.
Unlike a single-user single-railroad environment, the system of the present invention may encounter “conflicting” commands that attempt to write to and read from the devices of the model railroad. For example, the “conflicting” commands may inadvertently program the same device in an inappropriate manner, such as the locomotive to speed up to maximum and the locomotive to stop. In addition, a user that desires to read the status of the entire model railroad layout will monopolize the digital decoders and command stations for a substantial time, such as up to two hours, thereby preventing the enjoyment of the model railroad for the other users. Also, a user that programs an extensive number of devices will likewise monopolize the digital decoders and command stations for a substantial time thereby preventing the enjoyment of the model railroad for other users.
In order to implement a networked selective updating technique the present inventor determined that it is desirable to implement both a write cache and a read cache. The write cache contains those commands yet to be programmed by the digital command stations 18. Valid commands from each user are passed to a queue in the write cache. In the event of multiple commands from multiple users (depending on user permissions and security) or the same user for the same event or action, the write cache will concatenate the two commands into a single command to be programmed by the digital command stations 18. In the event of multiple commands from multiple users or the same user for different events or actions, the write cache will concatenate the two commands into a single command to be programmed by the digital command stations 18. The write cache may forward either of the commands, such as the last received command, to the digital command station. The users are updated with the actual command programmed by the digital command station, as necessary.
The read cache contains the state of the different devices of the model railroad. After a command has been written to a digital device and properly acknowledged, if necessary, the read cache is updated with the current state of the model railroad. In addition, the read cache is updated with the state of the model railroad when the registers of the devices of the model railroad are read. Prior to sending the commands to be executed by the digital command stations 18 the data in the write cache is compared against the data in the read cache. In the event that the data in the read cache indicates that the data in the write cache does not need to be programmed, the command is discarded. In contrast, if the data in the read cache indicates that the data in the write cache needs to be programmed, then the command is programmed by the digital command station. After programming the command by the digital command station the read cache is updated to reflect the change in the model railroad. As becomes apparent, the use of a write cache and a read cache permits a decrease in the number of registers that need to be programmed, thus speeding up the apparent operation of the model railroad to the operator.
The present inventor further determined that errors in the processing of the commands by the railroad and the initial unknown state of the model railroad should be taken into account for a robust system. In the event that an error is received in response to an attempt to program (or read) a device, then the state of the relevant data of the read cache is marked as unknown. The unknown state merely indicates that the state of the register has some ambiguity associated therewith. The unknown state may be removed by reading the current state of the relevant device or the data rewritten to the model railroad without an error occurring. In addition, if an error is received in response to an attempt to program (or read) a device, then the command may be re-transmitted to the digital command station in an attempt to program the device properly. If desirable, multiple commands may be automatically provided to the digital command stations to increase the likelihood of programming the appropriate registers. In addition, the initial state of a register is likewise marked with an unknown state until data becomes available regarding its state.
When sending the commands to be executed by the digital command stations 18 they are preferably first checked against the read cache, as previously mentioned. In the event that the read cache indicates that the state is unknown, such as upon initialization or an error, then the command should be sent to the digital command station because the state is not known. In this manner the state will at least become known, even if the data in the registers is not actually changed.
The present inventor further determined a particular set of data that is useful for a complete representation of the state of the registers of the devices of the model railroad.
An invalid representation of a register
A write only representation indicates that the register may not be read from. If this flag is set then a read error may not occur.
Over time the system constructs a set of representations of the model railroad devices and the model railroad itself indicating the invalid registers, read errors, and write errors which may increases the efficiently of programing and changing the states of the model railroad. This permit the system to avoid accessing particular registers where the result will likely be an error.
The present inventor came to the realization that the valid registers of particular devices is the same for the same device of the same or different model railroads. Further, the present inventor came to the realization that a template may be developed for each particular device that may be applied to the representations of the data to predetermine the valid registers. In addition, the template may also be used to set the read error and write error, if desired. The template may include any one or more of the following representations, such as invalid, in use, read error, write only, read dirty, read only, write error, and write dirty for the possible registers of the device. The predetermination of the state of each register of a particular device avoids the time consuming activity of receiving a significant number of errors and thus constructing the caches. It is to be noted that the actual read and write cache may be any suitable type of data structure.
Many model railroad systems include computer interfaces to attempt to mimic or otherwise emulate the operation of actual full-scale railroads.
For trains following each other, T&TO operation must rely upon time spacing and flag protection to keep each train a sufficient distance apart. For example, a train may not leave a station less than five minutes after the preceding train has departed. Unfortunately, there is no assurance that such spacing will be retained as the trains move along the line, so the flagman (rear brakeman) of a train slowing down or stopping will light and throw off a five-minute red flare which may not be passed by the next train while lit. If a train has to stop, a flagman trots back along the line with a red flag or lantern a sufficient distance to protect the train, and remains there until the train is ready to move at which time he is called back to the train. A flare and two track torpedoes provide protection as the flagman scrambles back and the train resumes speed. While this type of system works, it depends upon a series of human activities.
It is perfectly possible to operate a railroad safely without signals. The purpose of signal systems is not so much to increase safety as it is to step up the efficiency and capacity of the line in handling traffic. Nevertheless, it's convenient to discuss signal system principals in terms of three types of collisions that signals are designed to prevent, namely, rear-end, side-on, and head-on.
Block signal systems prevent a train from ramming the train ahead of it by dividing the main line into segments, otherwise known as blocks, and allowing only one train in a block at a time, with block signals indicating whether or not the block ahead is occupied. In many blocks, the signals are set by a human operator. Before clearing the signal, he must verify that any train which has previously entered the block is now clear of it, a written record is kept of the status of each block, and a prescribed procedure is used in communicating with the next operator. The degree to which a block frees up operation depends on whether distant signals (as shown in
The block signaling which does the most for increasing line capacity is automatic block signals (ABS), in which the signals are controlled by the trains themselves. The presence or absence of a train is determined by a track circuit. Invented by Dr. William Robinson in 1872, the track circuit's key feature is that it is fail-safe. As can be seen in
The track circuit is also an example of what is designated in railway signaling practice as a vital circuit, one which can give an unsafe indication if some of its components malfunction in certain ways. The track circuit is fail-safe, but it could still give a false clear indication should its relay stick in the closed or picked-up position. Vital circuit relays, therefore, are built to very stringent standards: they are large devices; rely on gravity (no springs) to drop their armature; and use special non-loading contacts which will not stick together if hit by a large surge of current (such as nearby lightning).
Getting a track circuit to be absolutely reliable is not a simple matter. The electrical leakage between the rails is considerable, and varies greatly with the seasons of the year and the weather. The joints and bolted-rail track are by-passed with bond wire to assure low resistance at all times, but the total resistance still varies. It is lower, for example, when cold weather shrinks the rails and they pull tightly on the track bolts or when hot weather expands to force the ends tightly together. Battery voltage is typically limited to one or two volts, requiring a fairly sensitive relay. Despite this, the direct current track circuit can be adjusted to do an excellent job and false-clears are extremely rare. The principal improvement in the basic circuit has been to use slowly-pulsed DC so that the relay drops out and must be picked up again continually when a block is unoccupied. This allows the use of a more sensitive relay which will detect a train, but additionally work in track circuits twice as long before leakage between the rails begins to threaten reliable relay operation. Referring to
From this standpoint, it is important to allow trains to move along without receiving any approach indications which will force them to slow down. This requires a train spacing of two block lengths, twice the stopping distance, since the signal can't clear until the train ahead is completely out of the second block. When fully loaded trains running at high speeds, with their stopping distances, block lengths must be long, and it is not possible to get enough trains over the line to produce appropriate revenue.
The three-block, four-indication signaling shown in
Some of the systems that are currently developed by different railroads are shown in
To take care of junctions where trains are diverted from one route to another, the signals must control train speed. The train traveling straight through must be able to travel at full speed. Diverging routes will require some limit, depending on the turnout members and the track curvature, and the signals must control train speed to match. One approach is to have signals indicate which route has been set up and cleared for the train. In the American approach of speed signaling, in which the signal indicates not where the train is going but rather what speed is allowed through the interlocking. If this is less than normal speed, distant signals must also give warning so the train can be brought down to the speed in time.
In the context of a model railroad the controller is operating a model railroad layout including an extensive amount of track, several locomotives (trains), and additional functionality such as switches. The movement of different objects, such as locomotives and entire trains, may be monitored by a set of sensors. The operator issues control commands from his computer console, such as in the form of permissions and class warrants for the time and track used. In the existing monolithic computer systems for model railroads a single operator from a single terminal may control the system effectively. Unfortunately, the present inventor has observed that in a multi-user environment where several clients are attempting to simultaneously control the same model railroad layout using their terminals, collisions periodically nevertheless occur. In addition, significant delay is observed between the issuance of a command and its eventual execution. The present inventor has determined that unlike full scale railroads where the track is controlled by a single dispatcher, the use of multiple dispatchers each having a different dispatcher console may result in conflicting information being sent to the railroad layout. In essence, the system is designed as a computer control system to implement commands but in no manner can the dispatcher consoles control the actions of users. For example, a user input may command that an event occur resulting in a crash. In addition, a user may override the block permissions or class warrants for the time and track used thereby causing a collision. In addition, two users may inadvertently send conflicting commands to the same or different trains thereby causing a collision. In such a system, each user is not aware of the intent and actions of other users aside from any feedback that may be displayed on their terminal. Unfortunately, the feedback to their dispatcher console may be delayed as the execution of commands issued by one or more users may take several seconds to several minutes to be executed.
One potential solution to the dilemma of managing several users' attempt to simultaneously control a single model railroad layout is to develop a software program that is operating on the server which observes what is occurring. In the event that the software program determines that a collision is imminent, a stop command is issued to the train overriding all other commands to avoid such a collision. However, once the collision is avoided the user may, if desired, override such a command thereby restarting the train and causing a collision. Accordingly, a software program that merely oversees the operation of track apart from the validation of commands to avoid imminent collisions is not a suitable solution for operating a model railroad in a multi-user distributed environment. The present inventor determined that prior validation is important because of the delay in executing commands on the model railroad and the potential for conflicting commands. In addition, a hardware throttle directly connected to the model railroad layout may override all such computer based commands thereby resulting in the collision. Also, this implementation provides a suitable security model to use for validation of user actions.
Referring to
The “asynchronous” receipt of commands together with a “synchronous” manner of validation and execution of commands from the multiple control panels 300 permits a simplified dispatcher controller 310 to be used together with a minimization of computer resources, such as com ports. In essence, commands are managed independently from the client program 14. Likewise, a centralized dispatcher controller 310 working in an “off-line” mode increases the likelihood that a series of commands that are executed will not be conflicting resulting in an error. This permits multiple model railroad enthusiasts to control the same model railroad in a safe and efficient manner. Such concerns regarding the interrelationships between multiple dispatchers does not occur in a dedicated non-distributed environment. When the command is received or validated all of the control panels 300 of the client programs 14 may likewise be updated to reflect the change. Alternatively, the controlling interface 16 may accept the command, validate it quickly by the dispatcher controller, and provide an acknowledgment to the client program 14. In this manner, the client program 14 will not require updating if the command is not valid. In a likewise manner, when a command is valid the control panel 300 of all client programs 14 should be updated to show the status of the model railroad 302.
A manual throttle 320 may likewise provide control over devices, such as the locomotive, on the model railroad 302. The commands issued by the manual throttle 320 may be passed first to the dispatcher controller 310 for validation in a similar manner to that of the client programs 14. Alternatively, commands from the manual throttle 320 may be directly passed to the model railroad 302 without first being validated by the dispatcher controller 302. After execution of commands by the external devices 18, a response will be provided to the controlling interface 16 which in response may check the suitability of the command, if desired. If the command violates the layout rules then a suitable correctional command is issued to the model railroad 302. If the command is valid then no. correctional command is necessary. In either case, the status of the model railroad 302 is passed to the client programs 14 (control panels 300).
As it can be observed, the event driven dispatcher controller 310 maintains the current status of the model railroad 302 so that accurate validation may be performed to minimize conflicting and potentially damaging commands. Depending on the particular implementation, the control panel 300 is updated in a suitable manner, but in most cases, the communication transport 12 is freed up prior to execution of the command by the model railroad 302.
The computer dispatcher may also be distributed across the network, if desired. In addition, the computer architecture described herein supports different computer interfaces at the client program 14.
The terms and expressions which have been employed in the foregoing specification are used therein as terms of description and not of limitation, and there is no intention, in the use of such terms and expressions, of excluding equivalents of the features shown and described or portions thereof, it being recognized that the scope of the invention is defined and limited only by the claims which follow.
This application is a continuation of U.S. patent application Ser. No. 10/989,815, filed Nov. 16, 2004, which is a continuation of U.S. patent application Ser. No. 10/713,476, filed Nov. 14, 2003, now U.S. Pat. No. 6,909,945, which is a continuation of U.S. patent application Ser. No. 09/311,936, filed May 14, 1999, now U.S. Pat. No. 6,676,089, which is a continuation of U.S. patent application Ser. No. 09/104,461, filed Jun. 24, 1998, now U.S. Pat. No. 6,065,406.
| Number | Name | Date | Kind |
|---|---|---|---|
| 6065406 | Katzer | May 2000 | A |
| 6530329 | Katzer | Mar 2003 | B2 |
| 6676089 | Katzer | Jan 2004 | B1 |
| 6909945 | Katzer | Jun 2005 | B2 |
| Number | Date | Country | |
|---|---|---|---|
| 20060241825 A1 | Oct 2006 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10989815 | Nov 2004 | US |
| Child | 11375794 | US | |
| Parent | 10713476 | Nov 2003 | US |
| Child | 10989815 | US | |
| Parent | 09311936 | May 1999 | US |
| Child | 10713476 | US | |
| Parent | 09104461 | Jun 1998 | US |
| Child | 09311936 | US |
