Data processing: artificial intelligence – Knowledge processing system – Knowledge representation and reasoning technique
Reexamination Certificate
1998-12-11
2001-02-20
Hafiz, Tariq R. (Department: 2762)
Data processing: artificial intelligence
Knowledge processing system
Knowledge representation and reasoning technique
Reexamination Certificate
active
06192356
ABSTRACT:
FIELD OF INVENTION
A method and system for automatically evaluating decisional rules containing tasks and conditions which must be fulfilled before the tasks can be performed wherein continuing iterative evaluations of whether the conditions are fulfilled are automatically performed and processing is resumed only after the conditions are fulfilled.
BACKGROUND OF INVENTION
Conventional computer-based systems generally have the capacity to quickly process many decisional rules of the type “IF [condition], then [perform task].” As processing occurs and the decisional rule is executed, if the condition is fulfilled, the task will be performed.
IF type logic statements, in fact, could fairly be called fundamental constructs in computer science-based technology and IF statements are standard language descriptors shared by most high level computer languages including Ada, Algol, BASIC, C, LISP, FORTRAN; and even software command languages such as dBASE III and SuperCalc. A system which has the ability to perform tasks based on the fulfillment of one or more conditions following an IF statement as that statement is executed is satisfactory for many applications.
Unfortunately, such systems do not include means to automatically evaluate whether the condition are fulfilled unless the decisional rule containing the IF command is itself currently being executed. Furthermore, sequential evaluation of the decisional rules is not automatic. Processing starts, the decisional rule is executed once, and processing then continues. To recursively evaluate the condition of the decisional rule, the decisional rule itself must be recursively executed. To accomplish this, computer programmers skilled in the art must program the system appropriately using “loops” so that recursive execution of the decisional rule occurs.
In many applications, however, such as in systems on board the space shuttle or space station, the astronaut or ground crew operators of such systems may not be sufficiently skilled in computer science disciplines to program the system to meet their needs. Furthermore, such programming is time consuming, cumbersome and complex in systems involving many decisional rules, subject to errors, and not ergonomically sound. In addition, such programming requires continuous execution in programming “loops” containing the decisional rules.
Most importantly, human beings think not in terms of recursive repetition of decisions whether to perform a given task; instead, they base their actions on when in time it would be appropriate to perform the task. The process of translating human thought into an abstract series of steps that could implement that thought has been the focus (and the success) of computer-based systems and artificial intelligence. The invention of high level programming languages was a major achievement in the development of a system which emulates not machine thinking, but human thinking. Recursive based evaluations of decisional rules, common in programming using high level languages, however, still emulate machine thinking. For example, imagine a given task is: at 30 seconds before launch, ignition of the shuttle rocket engine is to occur. This decision rule would look something like: “IF T=T
launch
−30, then ignite rocket engine.”
Prior art systems would execute the rule and in doing so, check once whether T=T
launch
−30 is true. If this condition is not true, execution ends. This prior art system could be forced to recursively execute this decisional rule until T=T
launch
−30 is true, (and then ignite the rocket engine), but the skill required to achieve this result, as explained above, is not elementary.
Since human beings have the capacity to easily function within the flow of time, a better system, and a system which more closely emulates human thinking, would automatically receive and process decisional rules based on the flow of time. The form of such a decisional rule might be: “When T=T
launch
−30, then ignite engine.” Unfortunately, no such system exists. As delineated above, the same result may be achieved, more or less, by individually designing the IF-based decisional rules of prior art systems, but achieving that result requires specialized training, and an often complicated design. Astronauts and others who often use such systems often lack specialized training, and moreover do not have sufficient time (while operating other systems on board a space craft) to engage in such design efforts.
Accordingly, there is a need for a system based on emulating human thinking and which therefore functions within the flow of time.
SUMMARY OF INVENTION
It is therefore an object of this invention to provide a system and method for executing a decisional rule which performs a task upon the fulfillment of a condition which provides automatic and continuing iterative evaluations of whether the condition is fulfilled until the condition is fulfilled at least once and which resumes any further processing only after the condition is fulfilled once.
It is a further object of this invention to provide such a system and method which does not require manual programming of the system in order to achieve continuing recursive evaluation of conditions contained within a decisional rule.
It is a further object of this invention to provide such, a system and method which operates within the flow of time.
It is a further object of this invention to provide such a system and method which allows as input constructs which more closely emulates human thinking thereby providing a sound ergonomic design.
It is a further object of this invention to provide such a system and method which automatically and iteratively evaluates whether the conditions of a decisional rule are fulfilled, thereby alleviating the need for manually providing additional decisional rules which force sequential evaluation in programming “loops.”
This invention results from the realization that a system and method for automatically executing decisional rules which perform tasks upon the fulfillment of conditions can be accomplished by automatically providing continuing iterative evaluations of whether the conditions are fulfilled until the conditions are fulfilled at least once and by allowing processing to resume only after the conditions are fulfilled at least once thereby permitting decisional rules to be in a form which more closely emulates human thinking and thereby also eliminating the need to manually force recursive execution of the decisional rules which would otherwise require knowledge of computer science disciplines and complex logic structures.
This invention feature a method for automatically evaluating a decisional rule containing a task and a condition which must be fulfilled before the task can be performed and for automatically performing the task when the condition is fulfilled. The method involves providing automatic and continuing iterative evaluations of whether the condition is fulfilled until the condition of the decisional rule is fulfilled. Then and only then is further processing allowed. Accordingly, the decisional rule is entered into computing means and compiled to parse and isolate the condition. In one embodiment automatic and continuing iterative evaluations are provided of whether the condition is fulfilled until the condition is fulfilled once, the task is then automatically performed when the condition is fulfilled once and further processing is resumed only after the condition is fulfilled once. In another embodiment, automatic and iterative evaluations of whether the condition is fulfilled are provided and the task is automatically performed whenever the condition is fulfilled. Further processing is resumed each time the condition is fulfilled. In still another embodiment the condition contains a time period which must be satisfied before the task can be performed. In that embodiment the task is performed automatically each time the time period is satisfied and processing is resumed each time the task is performed. T
Davis George
Hafiz Tariq R.
The Charles Stark Draper Laboratory Inc.
Wolf Greenfield & Sacks P.C.
LandOfFree
System and method for automatically executing decisional rules does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for automatically executing decisional rules, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for automatically executing decisional rules will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2597787