WO1999046703A1 - Optimization of a recipe for a spatial environment - Google Patents

Abstract

Systems and method are provided to optimize a recipe for a spatial environment. The method generally comprisess the steps of generating, analyzing a spatial database and devising a recipe. Generation of the spatial database involves the generation of both historic statements (110) and facts gathered from historic (112), present and future predicted events (114). Once the spatial database is generated, the facts are iteratively analyzed against the current statements (104) to see if they can or cannot be executed. If the facts can be executed, the facts are maintained as stored facts and analyzed for their economic feasibility. The determination of economic feasibility for the stored facts is accomplished by a iterative process similar to the determination of whether the facts can be executed. Once economic feasibility is determined, the recipe is devised. The method is optionally applied to the agricultural field for devising the fertilizer recipe, the irrigation recipe, and the split recipe.

Description

OPTIMIZATION OF A RECIPE FOR A SPATIAL ENVIRONMENT

CONTRACTUAL ORIGIN OF THE INVENTION

The United States has rights in this invention pursuant to Contract No. DE-AC07-94ID 13223 between the U.S. Department of Energy and Lockheed Martin Idaho Technologies Company.

RELATED APPLICATION

This application claims priority from provisional application S N 60/077,583, filed on March 10, 1998.

BACKGROUND OF THE INVENTION Field of the Invention

The present invention relates generally to systems and methods for optimizing a recipe for a spatial environment. More specifically, the present invention relates to systems and methods for fertilizing agricultural lands. Even more specifically, the present invention relates to systems and methods for generating a fertilizer recipe that is economically optimized for a particular agricultural field.

Relevant Technology

Since this invention has applicability to numerous disciplines, each having its own unique set of problems and shortcomings, for brevity, the following description of the relevant technology will pertain primarily to fertilizing and/or irrigating an agricultural field. 2

Soil types in an agricultural field are often categorized according to their relative proportions of sand, silt and clay. Although twelve generally accepted classifications of soil types exist, the classifications of sandy, loam and clay are the main classifications to which the others belong. Sandy soils are soils containing 85% or more of sand and a maximum of 10% of clay. As such, sandy soils have the least capacity to hold water and nutrients. Clay soils, on the other hand, have the greatest capacity to hold water and store nutrients. Thus, clay soils require the least amount of fertilization. Loam soils are a hybrid classification between sand and clay and consist generally of a friable mixture of varying proportions of clay, silt and sand. Although agricultural fields may not have all three of the main classifications, they typically contain at least two of the three with one classification being dominant. It would be unusual if an agricultural field only contained one soil type. Thus, since each agricultural field has varying soil types, each field has a variable requirement for applying fertilizer and water. The typical field is usually fertilized with uniform blends of fertilizers. At best, some fields are fertilized according to the variations in the soil types. In either event, the three most common nutrients in these fertilizers include nitrogen, potassium and phosphorous. However, applying these nutrients together causes a problem because these nutrients have one effect on the soils to which they are applied while simultaneously having another effect on each other. Moreover, fertilizers are often migrated away from the region where they were applied because of drainage, erosion and topographical characteristics of the field. Thus, it is problematic to provide a customized fertilizer blend to a field.

Additionally problematic is the reality of dispensing these fertilizer blends. Presently, the preferred fertilizer dispensing apparatus is a truck or tractor having a bulk fertilizer storage bin that can distribute large quantities of fertilizer to a field in a relatively short time. However, since soil types often change rapidly and drastically across a singular field, bulk distribution systems cause problems because they are unable to rapidly adjust to these soil changes. As a result, the bulk distribution systems frequently dispense properly blended fertilizers to the field at the improper location. 3

The agricultural field is also irrigated in a limited way. Typical irrigation methods include flood and sprinkler irrigation. Both methods attempt to deliver water to the agricultural field in a somewhat uniform manner. For example, sprinkler systems are often timed, or may rotate about a pivot at a specified speed. These irrigation methods do not account for the topography of the agricultural field and do not consider the current water content or evaporation rates of the agricultural field. As a result, some areas of the agricultural field may not receive enough water or the water may pond in other areas of the agricultural field. Neither condition is desirable for optimum field usage. These and other difficulties have been appreciated by the prior art. As such, various attempts have been undertaken to quantify the effects of distributing fertilizer blends and irrigation water across various topographies having wide- ranging erosion and drainage characteristics. The result has been detailed tabular data or charts. This data, however, has been identified as having assorted shortcomings.

One shortcoming is that the data is too "generalized" and not specific enough to an individual farmer's needs. Another shortcoming is that this data does not consider the history of a particular field. Yet the history of the field is an exceedingly consequential factor in the annual production of crops, which is one reason for rotating crops.

Agronomists, who are specialists in the agriculture branch of dealing with crop production and soil management, have also appreciated the foregoing problems. They have likewise attempted to quantify the effects of fertilizer, erosion and soils as it relates to crop production. Although considerably useful agronomy data has been generated, the data has weaknesses because it is frequently generated from closed environments, such as greenhouses and terrariums, and is not completely indicative of "real-world" growing environments.

Agronomists' data also has weakness because it is typically limited in scope. For example, agronomists frequently test soils to ascertain the major constituents of the soil while neglecting the minor constituents. This approach is 4 acceptable so long as the soil is not complex in constituency. For example, if the data conveys that wheat grows best at nitrogen levels of 39 parts per million (ppm) and the field is currently 30 ppm, the data recommends to add fertilizer having nitrogen in amounts to bring that portion of the field up 9 ppm. This recommendation, however, does not usually consider other pertinent information such as pH, lime, topography, irrigation, micronutrients, such as magnesium, boron, manganese and sulphur, and other complexities that serve to make each field unique. Thus, if that portion of the field has substantial amounts of boron, that field, altogether, might not even be a good place to grow wheat. But since the data says 39 ppm of nitrogen to grow wheat, then wheat growth is attempted in that field and fertilizer containing nitrogen is added. This example also serves to illustrate the problem that this data is usually crop-specific and is without capacity to adapt to other crops.

Within the prior art there is a farming discipline known as variable rate technology (NRT) that is used to apply variably blended fertilizer compositions to a field in a manner attempting to overcome the difficulties associated with bulk distribution systems. In U.S. Patents Νos. 4,700,895, 5,220,876, 5,355,815 and 5,689,418, all having common assignee Ag-Chem Equipment Co., Inc., of Minnesota, methods and apparatus are described to apply a fertilizer "prescription" unique to each particular agricultural field. In general, these patents combine to teach fertilization for a particular field by: (i) utilizing a soil map, particularized to the field, stored onboard a dispensing truck that is used to distribute the fertilizer; (ii) obtaining "real-time" soil samples from a soil sampler attached to the truck for supplementing and updating the soil map; and (iii) real- time variably adjusting the fertilizer blend from various nutrient bins stored upon the truck before distribution onto the field in order to "optimize" the fertilizer prescription.

These NRT dispensing trucks and methodologies are problematic for at least three reasons. One, the NRT dispensing truck system does not consider the economic worthiness of the fertilizer prescription. For example, if the soil sampler and the stored soil map, together with computer means, make a "real- 5 time" determination that a particular portion of the field has sandy soils and requires extensive fertilization, the truck will automatically distribute the necessary fertilizer blend according to the determination. But, if the determination to fertilize equates to a $15 bag of nitrogen and the current market price for the particular crop being fertilized will only yield an additional $10 of income, a costly economic decision has been made. In other words, a farming operation cannot expend more dollars in fertilizer than it receives when selling the harvested crops. Faulty economic decisions compounded over numerous fields over many years will eventually cause catastrophic financial ruin of a farming operation.

Additionally, if the determination requires the addition of fertilizer without considering other effects, such as topography and drainage, the fertilizer may be applied to an area of the field having substantial run-off. If this happens, the fertilizer may be washed away from its area of intended use before having an opportunity to contribute any nutrients to the soil. The run-off might even cause the fertilizers to flow into neighboring rivers and streams, for example, and contribute to an already burdensome environmental pollution problem.

Although not necessarily specific to these foregoing patents, fertilizer production is a colossal consumer of energy resources. As such, applying fertilizers to a poorly selected area of a field, i.e., sloping topography having high drainage, not only potentially increases environmental pollution, but causes society at large to suffer because the energy resources used to produce the fertilizer is unnecessarily expended if the fertilizer goes unused such as occurs when it is washed away. Such waste could be prevented if better determinations are made about when and where fertilizers are to be applied.

The foregoing NRT dispensing truck system does not consider the history of the particular field being fertilized. The VRT system utilizes a "snapshot" of the field taken at the particular time when soil samples are obtained by the truck and analyzed by the computer means. Although the calculations used by the computer to obtain the proper fertilizer prescription from the snapshot are not described in detail, nor taught in the patents because of a claim to proprietary and 6 trade secret information, the calculations are implied to be a function of the fertilizers themselves. This is because no information is taught in these calculations as to the topography, exposure, erosion, irrigation methods and other similar data. There is no mention of the field history in these calculations to ascertain whether the soil types exist as a function of previously applied fertilizer or whether the soil types are naturally occurring.

Once the NRT system prescribes a certain blend of fertilizers, the fertilizers are still distributed by a boom that is much larger in size than the soil sampler mechanism. What this means is that the soil still gets fertilized in bulk dosages not specific to the small sized area from which the soil sample was taken. Thus, the final fertilizer prescriptions are still largely informed guesses and economic optimization is again avoided.

Still another problem is that the presently available VRT dispensing trucks are limited to trucks having small numbers of bins, such as six, for storing and distributing fertilizer nutrients. Although six bins are reflective of the way that conventional fertilizers are sold, i.e., as a PΝK package indicating relative amounts of phosphorous, nitrogen and potassium, many more micronutrients are available for improving growing conditions, and ultimately crop yield. Yet these are largely overlooked by the NRT track technology. The prior art also largely overlooks other worthwhile information such as neighboring fields, irrigation methodologies, predicted market prices of various crops, predicted rainfalls and other similar data.

Accordingly, it is desirous to provide systems and methods for applying fertilizer to an agricultural field that is economically optimized and reflects all available past, present and future information useful in crop production. 7

OBJECTS AND SUMMARY OF THE INVENTION

It is, therefore, an object of one embodiment of the present invention to provide systems and methods for economically optimizing a fertilizer recipe for an agricultural field. It is a further object of one embodiment of the present invention to provide systems and methods for economically optimizing an irrigation recipe for an agricultural field.

It is another object of one embodiment of the present invention to provide systems and methods for optimizing any recipe in any spatial environment. It is a further object of one embodiment of the present invention to provide systems and methods for optimizing a recipe for a spatial environment utilizing historic, present and future information.

It is yet another object of one embodiment of the present invention to provide systems and methods for optimizing a recipe for an agricultural field that is non-crop specific and particular to the agricultural field.

It is still another object of one embodiment of the present invention to provide systems and methods for optimizing a recipe for an agricultural field utilizing expert systems to continually improve the optimization of the recipe.

It is still yet another object of one embodiment of the present invention to provide systems and methods that contribute to the reduction of environmental pollution and energy waste.

It is even yet another object of one embodiment of the present invention to provide a knowledge database processing system which promotes a user's judgment by combining relevant historical information together with current and predicted information to expand the knowledge of the database by feeding back a result of an actual or predicted event.

In accordance with the invention as embodied and broadly described herein, the foregoing and other objectives are achieved by providing systems and methods for determining an optimized recipe for a spatial environment. In general, the recipe is a set of instructions to be performed and the optimizing of the recipe is variable according to desired results such as time, finances or 8 resources. The spatial environment is any region relating to, occupying or having the characteristics of space to which variations occur, in some manner, throughout the space. Such a spatial environment includes an agricultural field having varying soil conditions throughout. The recipe may also be particular to the entire spatial environment or to just a portion thereof. In the context of the present invention, the systems and methods for optimizing a recipe for a spatial environment are described for economically optimizing a fertilizer recipe and/or an irrigation recipe for an agricultural field.

In a preferred embodiment, the method generally comprises the steps of generating a spatial database for the agricultural field, analyzing the spatial database and devising a recipe particular to the database and the agricultural field. Optionally, the method includes the step of applying and updating the recipe.

Generating the spatial database involves the generation of both statements and facts. Statements are characteristic about the field and are either historic or current. Historic and current statements are defined in relationship to one another. Facts are generated from the historic statements, and represent a condensed version of the historic statements. Preferably, the facts are generated by artificial intelligence routines.

Once the spatial database is generated, the facts are iteratively analyzed against the current statements to see if they can or cannot be executed. If the facts can be executed, the facts are maintained as stored facts and analyzed for their economic feasibility. If the facts cannot be executed, the facts are discarded from consideration in devising the recipe.

The economic feasibility of the stored facts is determined by a iterative process that considers whether the stored facts can or cannot be observed. The iterative process is similar to the determination of whether the facts can be executed. If the stored facts can be economically observed, the facts are used to determine the recipe and the stored facts having the greatest economical benefit are then selected for inclusion in the recipe. If the stored facts cannot be observed, the stored facts are discarded from consideration in devising the recipe. 9

Once the recipe is obtained, the farming operation has a set of instructions for fertilizing or for irrigating an agricultural field so that the crop yield can be economically optimized. The step of actually applying the fertilizer recipe or irrigation recipe to the spatial environment is optional. After the recipe is devised, the recipe is preferably updated to increase the knowledge of the spatial database and to improve the optimization thereof.

Updating the recipe includes systems and methods for both measuring results after the recipe has been applied to the field and for periodically providing updates according to various time intervals. Recipe updates are also described according to feedback means of updating the spatial database.

These and other objects and features of the present invention will become more fully apparent from the following description and appended claims, or may be learned by the practice of the invention as set forth hereinafter.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more fully understand the manner in which the above-recited and other advantages and objects of the invention are obtained, a more particular description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. Understanding that these drawings depict only typical embodiments of the invention and are not therefore to be considered to be limiting of its scope, the invention in its presently understood best mode for making and using the same will be described and explained with additional specificity and detail through the use of the accompanying drawings in which: Figure 1 is an exemplary computing system providing a suitable operating environment for the present invention;

Figure 2 is a flow diagram of the hierarchical operation of generating an optimized recipe for a spatial environment;

Figure 3 is a flow diagram of analyzing facts and devising a recipe for the spatial environment which is invoked by the routine of Figure 2; 10

Figures 4A to 4C are flow diagrams of alternative embodiments of updating the recipe for the spatial environment;

Figure 5 is an exemplary feedback loop for updating a recipe by updating the spatial database; and Figure 6 is a flow diagram of the operation of generating an optimized split recipe for a spatial environment.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present invention provides systems and methods for determining an optimized recipe for a spatial environment. As used herein, "spatial environment" is broad-ranging and means any region relating to, occupying or having the characteristics of space to which variations occur, in some manner, throughout the space. For example, spatial environments include, but are not limited to: an environmental waste site having various levels of hazardous and/or radioactive materials throughout the site; a bomb-target site having numerous exploded and unexploded ordinance devices scattered above and below the ground each having various levels of explosive matter therein; an underground natural gas reserve having natural gas stored in various quantities throughout the rock structure underlying the ground surface; an underwater volcano having various regions of gases and fissure points; and an agricultural site for growing trees, grains, fruits and vegetables.

It is preferred that the variations which occur throughout the spatial environment are classifiable in some manner to be able to obtain tangible data. As indicated, some classifiable variations include, radioactivity, explosive matter, volume, gas and structure. Other classifiable variations include, but are not limited to, finances, biology, chemicals, resource materials, topography, meteorology, geography, geology and other similar classifications. As used herein, these classifiable variations will be referred to as "modules" or "subsets" of the spatial environment. The total sum of the subsets or modules are together combined to make up the entirety of the spatial environment. However, when devising a recipe for a spatial environment it is not necessary to examine all 11 possible subsets and quite frequently it is advantageous to examine just one or a few of the subsets.

The spatial environment is also arranged as a compilation of many "spatial sites" or, for brevity, "sites" through which the modules exist. An example of a "site" at an environmental waste site might be a waste pit having numerous pockets of radioactive material with varying levels of radioactivity. A site might also be the reactor itself. Another site might be the various building structures housed on the grounds of the nuclear reactor. Thus, many spatial sites exist within the spatial environment and can vary in size and shape. Preferably, the sites represent a quantifiable size useful in devising a recipe as described below.

As used herein, "recipe" means a formula having various ingredients therein or a set of instructions to be performed. The recipe may be devised for either an individual module, a plurality of modules or for all the modules combined together. The formula or instructions provide a direction for achieving optimization.

As used herein, "optimize," and word derivatives thereof, means to attain the most favorable condition possible for a recipe. Exemplary most favorable conditions include, but are not limited to, time, finances, resources and other conditions being capable of optimization. An example of optimization in the event that radioactive materials have overly contaminated an environmental waste site include, a recipe to provide for the most "uncontaminated" clean-up of the materials, for the quickest clean-up or even the cheapest clean-up. It should be appreciated, however, that "local" most favored conditions are also attainable within a particular recipe. In the context of the radioactive clean-up for the nuclear reactor, local optimization might be the achievement of "uncontamination" for a singular 1 meter square plot, whereas the complete optimization is achievement of "uncontamination" for the entire reactor and not just one 1 meter square plot. Although defined as broad-ranging, in the context of the present invention, "optimization of a recipe for a spatial environment" will be specifically 12 exemplified in terms of a spatial environment being that of an agricultural field, such as that used to grow crops as part of a farming operation known commonly as "precision farming" or "precision agriculture." The "modules" of the agricultural field are numerous and exemplarily include, but are not limited to, a fertilizer schedule, an irrigation schedule, a herbicide schedule, a pesticide schedule, a seed-variety spacing schedule, an agricultural equipment schedule, an information management schedule and other similarly related agricultural topics useful in crop production. In the context of the present invention, the fertilizer schedule and the irrigation module will be the modules exemplarily used to describe the recipe.

The optimization of the fertilizer recipe and the irrigation recipe for the agricultural field will be exemplarily described in terms of an economic optimization. As used herein, "economic," and word derivatives thereof, means the operation of the agricultural field (hereinafter, for brevity may also be "field") in a careful, efficient and prudent manner such that financial waste or costs are minimized while benefits are maximized. "Economic" can accurately be represented by the difference between benefits and costs. Thus, economic optimization is achieved when the costs are relatively small when compared to the benefits. In the context of the module for a fertilizer schedule, such costs include, but are not limited to, the financial costs expended by the farming operation to obtain various blends of fertilizers. Benefits include, but are not limited to, the quantity of crops harvested or the price paid for the quantity of crops harvested. Thus, in the context of benefit/cost, "economic optimization" exemplary includes the lowest price spent for fertilizer in order to obtain the largest crop yield from that amount of fertilizer purchased.

It should be appreciated, however, that while additional money might be spent to obtain additional fertilizer or a higher grade of fertilizer and such expenditures might produce a more bountiful harvest, the difference between the benefits and costs might not be optimized for the farming operation. For example, consider that 100 bushels of wheat can be produced from a 1 acre plot without fertilizer and that a $15.00 bag of fertilizer will yield an extra 4 bushels 13 of wheat for that same acre. But, if wheat only commands $3.00 per bushel on the market, the farming operation would only receive an additional $12.00 for the extra 4 bushels. This is a net loss of $3.00 for that acre. Thus, the farming operation should take less yield in an economically optimized fertilizer recipe. In contrast, if a $20.00 bag of fertilizer will return an additional 7 bushels of wheat for that same acre at the same market price, a one dollar profit is potentially realized and the farming operation should invest in the extra fertilizer.

A similar analysis is applicable to the irrigation schedule. Economically, the decision of how much water to irrigate the agricultural field with and when to irrigate the agricultural field depends on the cost of irrigation versus the income from the projected yield. In accordance with the present invention, diagrams are used herein to illustrate either the structure or processing of embodiments used to implement the systems and method of the present invention. Using the diagrams in this manner to present the invention, however, should not be construed as limiting of its scope but merely as representative. As discussed in greater detail below, the embodiments of the present invention may comprise a special purpose or general purpose computer comprising various computer hardware.

Embodiments also within the scope of the present invention include computer readable media having executable instructions or data fields stored thereon. Such computer readable media can be any available media which can be accessed by a general purpose or special purpose computer. By way of example, and not limitation, such computer readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic disk storage devices, or any other medium which can be used to store the desired executable instructions or data fields and which can be accessed by a general purpose or special purpose computer. Combinations of the above should also be included within the scope of computer readable media. Executable instructions exemplarily comprise instructions and data which cause a general purpose computer, special purpose computer, or special purpose processing device to perform a certain function or group of functions. 14

Further contemplated are a hierarchy of storage devices that are available to the system. Such storage devices may comprise any number or type of storage media including, but not limited to, high-end, high-throughput magnetic disks, one or more normal disks, optical disks, jukeboxes of optical disks, tape silos, and/or collections of tapes or other storage devices that are stored off-line. In general, however, the various storage devices may be partitioned into two basic categories. The first category is local storage which contains information that is locally available to the computer system. The second category is remote storage which includes any type of storage device that contains information that is not locally available to a computer system. While the line between these two categories of devices may not be well defined, in general, local storage has a relatively quick access time and is used to store frequently accessed data, while remote storage has a much longer access time and is used to store data that is accessed less frequently. The capacity of remote storage is also typically an order of magnitude larger than the capacity of local storage.

Figure 1 and the following discussion are intended to provide a brief, general description of a suitable computing environment in which the invention may be implemented. Although not required, the invention will be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. Moreover, those skilled in the art will appreciate that the invention may be practiced with other computer system configurations, including hand-held devices, multi-processor systems, microprocessor-based or programmable consumer electronics, network

PCs, minicomputers, computer clusters mainframe computers, and the like. The invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices. 15

With reference to Figure 1, an exemplary system for implementing the invention includes a general purpose computing device in the form of a conventional computer 20, including a processing unit 21, a system memory 22, and a system bus 23 that couples various system components including the system memory to the processing unit 21. The system bus 23 may be any of several types of bus structures including a memory bus or memory controller, a peripheral bus, and a local bus using any of a variety of bus architectures. The system memory includes read only memory (ROM) 24 and random access memory (RAM) 25. A basic input/output system (BIOS) 26, containing the basic routines that help to transfer information between elements within the computer 20, such as during start-up, may be stored in ROM 24. The computer 20 may also include a magnetic hard disk drive 27 for reading from and writing to a hard disk, not shown, a magnetic disk drive 28 for reading from or writing to a removable magnetic disk 29, and an optical disk drive 30 for reading from or writing to removable optical disk 31 such as a CD-ROM or other optical media. The hard disk drive 27, magnetic disk drive 28, and optical disk drive 30 are connected to the system bus 23 by a hard disk drive interface 32, a magnetic disk drive-interface 33, and an optical drive interface 34, respectively. The drives and their associated computer-readable media provide nonvolatile storage of computer readable instructions, data structures, program modules and other data for the computer 20. Although the exemplary environment described herein employs a hard disk, a removable magnetic disk 29 and a removable optical disk 31, it should be appreciated by those skilled in the art that other types of computer readable media which can store data that is accessible by a computer, such as magnetic cassettes, flash memory cards, digital video disks, Bernoulli cartridges, random access memories (RAMs), read only memories (ROM), and the like, may also be used in the exemplary operating environment.

A number of program modules may be stored on the hard disk, magnetic disk 29, optical disk 31, ROM 24 or RAM 25, including an operating system 35, one or more application programs 36, other program modules 37, and program data 38. A user may enter commands and information into the computer 20 16 through input devices such as a keyboard 40 and pointing device 42. Other input devices (not shown) may include a microphone, joy stick, game pad, satellite dish, scanner, or the like. These and other input devices are often connected to the processing unit 21 through a serial port interface 46 that is coupled to system bus 23, but may be connected by other interfaces, such as a parallel port, game port or a universal serial bus (USB). A monitor 47 or other type of display device is also connected to system bus 23 via an interface, such as video adapter 48. In addition to the monitor, computers typically include other peripheral output devices (not shown), such as speakers and printers. The computer 20 may operate in a networked environment using logical connections to one or more remote computers, such as remote computer 49. Remote computer 49 may be a personal computer, a server, a router, a network PC, a peer device or other common network node, and typically includes many or all of the elements described above relative to the computer 20, although only a memory storage device 50 has been illustrated in Figure 1. The logical connections depicted in Figure 1 include a local area network (LAN) 51 and a wide area network (WAN) 52 that are presented here by way of example and not limitation. Such networking environments are commonplace in offices enterprise-wide computer networks, intranets and the Internet. When used in a LAN networking environment, the computer 20 is connected to the local network 51 through a network interface or adapter 53. When used in a WAN networking environment, the computer 20 typically includes a modem 54 or other means for establishing communications over the wide area network 52, such as the Internet. The modem 54, which may be internal or external, is connected to the system bus 23 via the serial port interface

46. In a networked environment, program modules depicted relative to the computer 20, or portions thereof, may be stored in the remote memory storage device. It will be appreciated that the network connections shown are exemplary and other means of establishing a communications link between the computers may be used. 17

With reference to Figure 2, a flow diagram for a hierarchical method of optimizing a recipe for a spatial environment is depicted generally as 100. Preferably, the flow diagram is implemented as the computer-executable instructions of a computer-readable medium exemplarily described in the context of the computing operating environment of Figure 1. Specifically, the flow diagram will be described as a method for economically optimizing the fertilizer schedule recipe and the irrigation schedule recipe of an agricultural field. In general, the method first comprises the generation of a database for the spatial environment at step 102, hereinafter referred to as a spatial database. Secondly, the method comprises the analysis of the spatial database at step 104. Thirdly, the method comprises the devising of a recipe 106 and optionally comprises the updating of the recipe at step 108.

The generation of the spatial database 102 comprises the characterization of unknowns, variables and constraints for the spatial environment that is to be managed, i.e., the agricultural field. In this embodiment, this includes the generation of "statements," "historic" or "current," and the generation of "facts."

As used herein, "statements" are individual descriptions of one characterization dataset that describes conditions that occur, have occurred or will occur for a "spatial site." An example of a spatial site in an agricultural field, might be a substantially square plot having about 1 meter sides. A site might also be larger and be about 70 feet per side. The site might even be defined as being 1 full acre. The site having about 70 feet per side is a presently useful site description because, as discussed in the background section, the current generation of NRT fertilizer truck/ tractor spreaders for precision farming have booms to distribute the fertilizer. Those booms are about 70 feet in length.

An example of a statement for a fertilizer module or an irrigation module is the set of soil nutrient parameter values as measured at one of the particular spatial sites. In particular, a statement might read: (i) at the first spatial site, the soil nutrients are 42 ppm nitrogen, 32 ppm phosphorous and 21 ppm potassium; or (ii) at the second spatial site the soil nutrients are 44 ppm nitrogen, 32 ppm phosphorous and 19 ppm potassium. Thus, it should be appreciated that for a 18 typically sized agricultural field of about 200 acres having sites that are substantially square and about 70 feet per side, thousands of statements are generated for the soil types alone. Other examples of statements include, but are not limited to: the soil chemistry; the soil concentrations of micronutrients such as boron, sulphur, and manganese; the soil "physics" such as soil leeching and water capacity; the parameters of the topography of each site; the terrain slope of each site; the drainage of each site, and so on.

The statements utilized by the fertilizer module may overlap with the statements utilized by the irrigation module and in some cases may be the same statements. The difference is how the statements are interpreted. The statements for the fertilizer module are used to determine how fertilizer should be applied to the spatial environment, whereas the statements for the irrigation module are used to determine how the spatial environment should be irrigated. For example, a statement quantifying the amount of nitrogen in the soil may be used by the fertilizer module to devise a recipe calling for more nitrogen. The same statement may be used by the irrigation module to limit the quantity of water applied to the spatial environment to limit the amount of nitrogen that is washed away through drainage of the spatial environment.

Many statements quantify the chemicals in a field or describe weather patterns of a particular area. Statements, however, are not limited to descriptions of data that accumulate over time or that can be measured. Statements may include crop models and other scientific information. In some instances, the statements containing scientific data may be viewed as a constraint. For example, the statements of an agricultural field may indicate that nitrogen should be added to the soil or that more water is needed. A crop model or scientific data, however may indicate that this particular soil can only accommodate a certain level of nitrogen. This may indicate that it is economically advantageous to add less nitrogen to the fertilizer recipe. Consider an example where the statements indicate that a substantial amount of phosphorous is lacking in the soil. Scientific data shows that the amount of phosphorous that the soil can absorb is limited.

Attempting to apply more phosphorous than the soil can absorb is not only 19 economically wasteful, but potentially harmful to the crop. In other words, statements can contain scientific data such as crop models as well as data actually measured and recorded as well as other types of relevant information. In addition to crop models, scientific information can include plant physiology and growth rates.

Statements may also be described as a set or subset of maps. For example, each agricultural field has many maps associated with it. One map may quantify the nitrogen in the soil while another map describes the type of soil. Other maps may be indicitative of weather patterns or evaporation rates or may contain crop models. The point is that each field has a plurality of maps containing information related to the agricultural field. Each map is divided into spatial sites or sections. The sum of all the information in each map for each spatial site is a statement. In other words a statement for a spatial site is the sum of the many maps or layers of data. The spatial database contains all the maps as well as the statements those maps generate.

Statements such as the above can be obtained or developed from many and diverse sources. Such sources include, but are not limited to: sensors, both remote and in situ; maps; charts; meteorological monitoring; wind calculations; temperature observations and predictions; relative humidity; crop models; and other related sources. Note that each of these sources can produce at least one map.

Statements are divided into at least two types, historic or current. An historic statement is a statement from a "previously" occurring condition. A current statement is a statement that describes "presently" existing conditions. It should be appreciated, however, that the dichotomy between previously and presently occurring conditions are largely defined in relation to one another. For example, if the soil at a spatial site was tested during the last growing season and was about 41 ppm nitrogen and about 20 ppm potassium, this would be an historic statement in the spatial database. If the soil is tested during the present growing season and is revealed to contain about 46 ppm nitrogen and about 26 ppm potassium, this is a current statement. Yet, in about one week, the soil 20 composition might actually be different than its presently tested values. But because of the impracticality of testing soil at numerous sites all over the agricultural field every few days or hours, current statements remain "relative" to historic statements. Thus, although the current statement is now about one week old and has actually been "previously" recorded, the current statement will remain as the "current statement" and the sample from the last growing season will remain as the "historic" statement. This is not to say, however, that if the soil is tested again during the middle of the growing season, for example, and different soil composition results are obtained, that the current statement would necessarily remain "current." In that situation, an update would occur as described below.

It should also be appreciated, that "current" statements might actually be from conditions having not yet occurred. For example, the expected price of a bushel of wheat for the present growing season might be expected at $3.25 a bushel. Although the wheat has not been sold yet, yet alone planted or harvested, the expected price of wheat may still be entered as a "current" statement. Thus, "current statements" comprise presently occurring conditions and predictions of future occurring events.

At step 110, the historic statements are generated and entered into the spatial database by a user in computing means well known in the art. Some historic statements for a fertilizer schedule recipe as well as an irrigation recipe include, but are not limited to: (i) tabular data of various fertilizer compositions according to brand, according to price, according to nutrients; (ii) soil type classifications from previous growing seasons according to each spatial site in the agricultural field; (iii) crop yields from previous growing seasons, in quantity and in price; (iv) previous rainfall and water irrigation amounts affecting the previous growing season; (v) information obtained from nearby agricultural fields for previous growing seasons; (vi) topography; and any other quantifiable information meeting the "relative" definition of historic statement. One presently preferred method of assisting in the generation of statements for the soil types in an agricultural field includes the classifying of soil 21 types that have been collected as samples at some grid pattern on the field. Each point of the grid pattern is a spatial site from which soil is collected and then analyzed so that statements can be generated. While the "best" grid pattern to be used for soil collection is disputed amongst scholars and agronomists, gridding arrangements are still agreed upon as the preferred method for obtaining soil samples in precision agriculture. Once samples are obtained, commercial products are available to assist in predicting how the nutrients in the soil are spatially arranged across the entire field even from areas of the field where no soil samples were collected. One preferred product is a Geographic Information System (GIS) software product sold commercially under the name Arc/Info by Environmental Systems Research Institute (ESRI) of California. Using this GIS product, soil data collected from the grid pattern can be interpolated until a map of the soil nutrients is produced for the entire field.

Once the historic statements are generated at step 110, the facts are generated from those historic statements at step 112. As used herein, "facts" are a set of descriptors condensed from the "knowledge" of the spatial database provided by the historic statements. The facts summarize the limits bounding some set of conditions. An example of a fact is a descriptor relating the quantity of crop yield for a given historic statement. Thus, for a plurality of historic statements generated for the numerous recorded soil types at each of the spatial sites throughout the agricultural field, the fact might read that 105 bushels of wheat per acre were produced from these soil types. Other facts are similarly generated from the historic statements to represent the knowledge of the spatial data base in an abridged version. In a preferred embodiment, facts are generated from the historic statements using artificial intelligence (AI) routines. These AI routines are known to those skilled in the art and are exemplified in various commercial embodiments. As such, the AI routines and methodologies are not discussed herein in detail. The first time the spatial environment is having a recipe devised therefor, there might not be any data available that can be used to generate historic 22 statements. Advantageously, the step of generating facts may alternatively be performed without having any historic statements to begin with. These facts generated are referred to as "generic" or "fundamental facts" and may include generally accepted data such as mathematical theorems, chemical reactions, electrical theories, physic equations and the like. Fundamental facts may also include things such as tabular data used currently by agronomists.

It would be rare if no historic statements at all could be generated because data may be accumulated from numerous and wide ranging sources. Thus, fundamental facts can be generated from such other sources like neighboring or regional spatial environments, weather history, newspaper reports eyewitness accounts and so on. In fact, data from neighboring spatial environments, such as nearby agricultural fields, would not only be beneficial as a base point but would certainly be directly related and relevant. For example, nearby agricultural fields would be able to furnish rainfall measurements and climatology information since nearby fields experience the substantially same precipitation and weather throughout the course of a singular growing season.

At step 114, currents statements are generated for inclusion in the spatial database. Current statements for a fertilizer schedule recipe module or an irrigation schedule recipe module include, but are not limited to: (i) the expected market price or a contractual price for a given crop; (ii) the expected or current market price for given fertilizer blends and various micronutrients; (iii) soil type classifications from the present growing season according to each spatial site in the agricultural field; (iv) topographical information about each spatial site; (v) water amounts received and predicted for each of the spatial sites; (vi) information obtained from nearby agricultural fields for the present growing season; (vii) climate and any other quantifiable information meeting the "relative" definition of current statement.

Although step 114 of generating the current statements is illustrated as sequentially following the generation of the facts at step 112, it should be appreciated that the current statements may be generated at any time during the development of the spatial database. The step 114 of generating current 23 statements may precede the generation of the historic statements or facts or may even be generated periodically throughout the development of the spatial database.

Once the spatial database is generated, the step of analysis is performed. As described in further detail below, the analysis at step 104 occurs generally by making decisions about the facts in light of the current statements. Although not depicted, the analysis 104 occurs for a particular module(s) of the spatial environment as entered by a user from the exemplary operating environment. Also user entered is the type of optimization that is desired for that particular module, i.e., time, resources, finances, etc. Again, in this embodiment, the modules are the fertilizer schedule and the irrigation module and the optimization for both modules is economics. Thereafter, once the analysis occurs, the recipe is devised for the module or modules of the spatial environment at step 106.

It should be appreciated, however, that the recipe may be a provided as a piecemeal or fragmented recipe over time. An example of a piecemeal recipe includes a fertilizer schedule or irrigation schedule optimized for price before the growing season and during the middle of the growing season. Such piecemeal recipe determination is best accomplished when the recipe is updated 108. Although described in alternative embodiments below, the updating of the recipe is generally the attainment of more data so that more statements and facts can be generated. Ultimately, this provides a larger spatial database, hence more knowledge, from which the recipe can be improved and better optimized. This is because, as is known, AI routines, expert systems, neural net trains and other similar systems like those having application in the present invention are all improved as knowledge is gained and as trial and error is recognized. Updating

108, however, is not a requirement for generating a recipe, but is merely preferred.

Another method of devising or developing the recipe is a split application. Split application recipes add a time element to the formulation of the recipe. The time element can be a statement in time. In other words, each spatial site has, in addition to all the statements described above, a temporal statement or map. For 24 example, consider a fertilizer recipe or irrigation recipe developed for an agricultural field used to grow potatoes. The recipe is developed at some point early in the growing season. While the statements described above enable a user to develop an optimal fertilizer recipe, the recipe cannot anticipate future events. By way of example and not limitation, if the recipe was developed for an agricultural field having an average of 90 growing degree days, the ultimate yield of the agricultural field is affected if there are more or less growing degree days. If an irrigation recipe is developed for an agricultural field having 6 inches of average rainfall and the agricultural field receives 10 inches of rain, then the irrigation recipe will not result in optimum economic return. In other words, recipes cannot compensate for future occurrences or future statements.

Split application recipes is a method of eliminating or reducing the effect of future statements. In general, a split application recipe is illustrated as method 200 in Figure 6. In step 202, a recipe is devised for a first stage. Step 202 includes the steps of: generating facts and statements for the spatial environment; analyzing the facts to determine whether the facts can be complied with or executed; and devising a recipe for the facts that are determined to be feasible. Once the first stage is complete, which may be indicated by a passage of an allotted time or by information supplied by sensors, step 204 devises a recipe for a next stage. Devising the recipe for the next stage is performed in a manner similar to the steps listed for the first stage. Step 206 ensures that this process is repeated until all split application recipes have been devised. The split application recipes devised by method 200 are usually separated temporally. The length of time between split applications may be influenced by economic factors. For crops, a preferred time period is approximately four weeks.

By way of example and not limitation, method 200 in Figure 6 may be illustrated in the context of a split application fertilizer recipe for use in growing potatoes. In this example, a split application recipe is devised in using the method described herein rather than devising a recipe for the entire growing season. A split application recipe is divided into various stages, which may correspond to a period of time or be related to plant growth. In this example, the 25 split application recipe is guided by stages corresponding to plant growth. The first stage for this example is the emergence of the potato plant. Step 202 devises a split application recipe guided by this first stage. Note that, in this case, the temporal period between the application of the split application fertilizer recipe and the emergence of the potato plant is not a set time period, but is based on the growth of the potato plant. Once the plant has emerged, the temporal occurrences can be taken into account in devising the next split application recipe. In other words, a temporal statement is now available to use in the computation of the next recipe. The next split application recipe in step 204 may be devised to cover the ground and close the rows, which essentially means that the split application recipe is intended to foster leaf growth such that the potato plants receive as much solar energy as possible. This is called closing the rows because the ground underneath the potato plants, ideally, cannot be seen. In step 206, another split application is needed to promote tuber growth. Thus step 204 is again executed by devising a next split application recipe intended to promote the reproductive phase of the plant. In other words, the energy of the plant should be directed toward tuber growth rather than leaf growth. Split application fertilizer recipes enable a user to respond to conditions that occur over time. In that sense, split application takes advantage of temporal statements. In other words, one embodiment of a temporal statement contains information about the time between split application recipes.

Remote sensing is important when using split application recipes and is especially important with regard to the irrigation schedule. In some embodiments of split application recipes, remote sensing determines when the next split application is to occur. Remote sensors which indicate the rate of evaporation or the water content of the soil provide important information that is used to develop the next recipe. With regard to an irrigation schedule, split application recipes enable a user to account for dry weather as well as wet weather. This is accomplished in the temporal statement, which may indicate rainfall or degree growing days. 26

Another important factor to consider when using split application is cost. While the cost for the irrigation schedule may not be excessive, the fertilizer schedule is more sensitive to cost. A weekly split application may produce optimum crop yield, but the temporal cell or statement is typically set at four weeks due to economic constraints for the fertilizer schedule. Split application, as well as single application fertilizer schedules can be developed regardless of how the fertilizer is delivered to the crop. Split application is not limited to fertilizer and irrigation schedules, but may be implemented with others including herbicide and pesticide applications. Another consideration that impacts the development of a fertilizer recipe or irrigation recipe is the global restraint of resources. By way of example and not limitation, a government or other entity may limit the amount of fertilizer that may be used or the amount of water may be very limited, especially in a drought. Under these types of circumstances, the methods of the present invention consider aggregate constraints when formulating the recipe.

In order to take into account the aggregate constraints, it is helpful to know the current statements of the agricultural field. While the current statements are not necessary in order to take aggregate constraints into account, they are preferred. Devising a recipe under aggregate constraints can be done using historic statements alone, but the recipe may not be optimum because current statements have not been considered. In the absence of aggregate constraints, a recipe is formulated that will result in optimal yield for that spatial site. This is accomplished as will be described in more detail below, by selecting the fact for each spatial site that will maximize the economic return and devising a recipe that will realize that fact. By way of example and not limitation, the fact that will result in a maximum economic return is to add a certain quantity of nitrogen to the soil, then the recipe is designed such that that amount of nitrogen will be applied to that spatial site.

In the presence of aggregate constraints, the facts are analyzed to produce an optimal recipe for the spatial environment rather than an optimal recipe for each spatial site within the spatial environment. By way of example and not 27 limitation, an agricultural field in need of nitrogen in an area where the amount of nitrogen that may be applied as fertilizer is regulated is subject to an aggregate constraint. In this case, the user or farmer is in a situation where more nitrogen is needed than can be applied because of regulations or other reasons. In order to maximize the return, the fertilizer recipe may indicate that nitrogen be applied in spatial sites having little to no drainage, or that nitrogen be applied in greater quantity to specific types of soil. In this manner, the economic return is maximized. A similar analysis can be performed where the amount of water available for irrigation is limited. When limited water is available, the irrigation recipe will be allocated to the various spatial sites such that economic return is maximized. In other words, the facts that can be executed for each spatial site are analyzed to produce a set of facts that generate a maximum return. The facts are analyzed using analytical tools well known in the art to produce a recipe that will maximize the economic return of the agricultural field while complying with aggregate constraints. When an agricultural field is under aggregate constraints, the recipe will produce an optimum return for the field, but the recipe for each spatial site within the agricultural field may or may not be formulated to produce a maximum return for that spatial site.

In Figure 3, the combined steps of analysis 104 and devising a recipe 106 for a spatial environment are more fully illustrated in the context of economically optimizing a fertilizer recipe and an irrigation recipe for an agricultural field. In Figure 3, the combined analysis and recipe determination is depicted generally as method 116.

In this embodiment, the analysis begins by obtaining a fact from the spatial database at step 118. A preliminary determination about the obtained fact is made at step 120. At step 120 the fact is examined against the backdrop of the current statements to see if the fact can or cannot be executed.

If the fact cannot be executed, the fact is discarded 122. An example of non-compliance and discarding is as follows: if the fact states "keep nitrogen below 42 ppm for wheat production" and a current statement indicates that the soil at a particular site in a field for growing wheat is determined to be 46 ppm 28 nitrogen, the fact cannot be executed; the fact is then discarded at step 122. Discarding of the facts in this manner eliminates superfluous data from being considered when the recipe is being devised. Thus, the final recipe is free from extraneous data. Once discarded, the method 116 then ascertains whether other facts are available 124. If so, the steps are iteratively processed beginning at 118 until all facts have been examined.

If the fact can be executed at step 120, step 126 is invoked. Step 126 isolates facts that can be executed and groups these facts together as "stored facts." Stored facts, however, should not be deemed to be actually stored as part of the exemplary operating environment on remote or local storage devices.

Although the facts could actually be stored, the "stored facts" are merely a means for describing the computer-executable instructions for isolating and/ or maintaining facts until such time as they are further considered as part of the recipe. As with discarded facts, once a fact is stored at step 126, the method 116 again ascertains whether other facts are available 124 for determining compliance. If more facts are available, the steps are iteratively processed until all facts have been exhausted. If no more facts are available, or once all facts have been examined to see if they can be executed, a similar iterative process is invoked for the stored facts. At step 128, a stored fact is obtained for a determination at step 130 to see whether the stored fact can or cannot be economically observed.

In the context of the costs and benefits, economic observation includes the determination of obtaining the optimized difference between benefits and costs. Thus, for the 1 acre farm plot that can produce 100 bushels of wheat without fertilizer and 104 bushels by using a $15.00 bag of fertilizer, the stored fact regarding the purchase of more fertilizer can be economically observed so long as the market price per bushel of wheat will yield a profit for those 4 extra bushels of more than $15.00. In other words, the stored fact is economically observable if the market price is more than $3.75 per bushel. If the market price is such that the price paid for the 4 extra bushels is less than $15.00, the stored fact cannot be 2 9 economically observed. If the market price is such that the price paid for the extra 4 bushels is exactly $15.00, which equals the price of the bag of fertilizer to achieve those extra bushels, the computer-executable instructions can be arranged to either include the stored fact in the recipe, exclude it or leave it up to the farming operation to decide.

If the stored fact cannot be economically observed, the stored fact is discarded at step 132. Similar to the facts, this method iteratively examines the other stored facts until no more stored facts remain. This iteration begins by determining whether other stored facts are available at step 134. If yes, another stored fact is obtained at step 128.

If the stored fact can be economically observed at step 130, the stored fact is included in the recipe for that module 136. As the method 116 iterates this process of examining stored facts, additional stored facts are added to the recipe. Each stored fact in the recipe may be referred to as an ingredient or an instruction to be performed.

It should be appreciated that some stored facts might be better for optimization than other stored facts. For example, one stored fact might indicate that applications of nitrogen will yield an extra 3 bushels of crop per acre. Another stored fact might indicate that additions of phosphorous will yield an extra 3.5 bushels per acre. Although both are economically feasible, the addition of phosphorous is "more economical" than the addition of nitrogen. In instances such as these, the most optimized stored fact overwrites the less optimized stored fact.

Once all stored facts have been examined and no other stored facts are available 134, method 116 ends by providing the recipe to the user 138 so that each ingredient may be applied to the agricultural field to achieve economic optimization of the fertilizer schedule.

It should be appreciated that, while method 116 has been described as obtaining a single fact and going through the steps of determining compliance, still other routines are available for cycling through all of the facts to determine whether they comply or not. Such other routines include, but are not limited to, 30 multiple looping schemes for simultaneously examining a plurality of facts, assigning a hierarchy of importance to the facts to which only the most important facts are iteratively examined and other similar routines. These routines are also embraced within the scope of the present invention. Method 116 also produces an optimized recipe for an irrigation schedule.

Step 118 obtains facts from the statements related to the irrigation schedule rather than the statements related to a fertilization schedule. All other steps of method 116 operate in a similar manner. The difference between method 116 as applied to an irrigation schedule as opposed to a fertilizer schedule is that one recipe is for the application of fertilizer and the other is for water. Also, the statements, from which the facts are derived may be different. In other words, statements that are relevant to a fertilizer schedule may or may not be relevant for an irrigation schedule. It is noted, however, that many of the statements will be identical. For example, both schedules most likely have statements concerning the topography, yield, drainage, and nitrogen content. Note however, that the facts derived from a similar set of statements are most likely different because of the different type of schedule. This is true for any type of schedule.

As indicated previously, updating the recipe 108 develops a more knowledgeable spatial database which ultimately improves optimization. With reference to Figures 4 A through 4C, some exemplary alternative embodiments for updating the recipe are illustrated.

In Figure 4A, the initial step of updating recipe 108 begins by applying the recipe to spatial environment 140. For the fertilizer schedule this might mean applying various fertilizer blends to the field at each of the spatial sites. Once applied, a chosen variable is measured 142. An example of a measuring a chosen variable is to take another soil sample to determine the effect the fertilizer had on the nutrients contained therein. For example, a chosen variable might be nitrogen and before application of the fertilizer recipe to the field a particular spatial site contained 42 ppm before the growing season and contained 44 ppm during the middle of the growing season when the nitrogen content was measured again.

This change in nitrogen can then be used to provide additional knowledge for the 31 spatial database. Thus, the chosen variable is fed back into the spatial database at step 144.

This feedback of the chosen variable into the spatial database is more fully illustrated by the block diagram of Figure 5. In Figure 5, the chosen variable or variable 146 is obtained at the output of the devising of the recipe as described at step 106 and fed back into the spatial database, illustrated by dashed line 148. The chosen variable is fed back as either a statement or a fact. In particular, the variable 146 can be used to generate an additional historic statement 150, to generate an additional fact 152 or to generate an additional current statement 154. From this figure, it should also be appreciated that the spatial database is not sequentially generated, as previously described, but is generated at various time intervals by a user 156 supplying input for the generation of either the historic statements 158 or the current statements 160.

Another alternative embodiment for updating the recipe 108 is illustrated in Figure 4B. This update begins by measuring some parameter in the spatial environment 162. The parameter is similar to a chosen variable, yet, in this embodiment, it is not necessary to have applied the recipe to the spatial environment. An example of a parameter is the amount of rainfall recorded in the spatial environment. The next step 164 is to translate what was known as a current statement into what is now known as an historic statement. As an illustration of this, consider if voluminous amounts of rainfall fell on a field and leeched the soil of valuable nutrients. This rainfall would impact upon the optimization of the fertilizer schedule. Accordingly, the recipe needs updating. Now the current facts indicating the present state of the nutrients in the soil is inaccurate due to the rainfall and the current statements are no longer valid. But instead of discarding the information, the database can be enlarged by translating what is current into something that is historic. Thereafter, since the historic statements have been enlarged, the facts should be updated at step 166.

A third alternative embodiment of updating the recipe 108 includes a combination of the steps performed in both of the alternative embodiments of

Figures 4A and 4B. In Figure 4C, the fertilizer recipe is applied to the spatial 32 environment 140, a chosen variable such as nitrogen is measured 142 and the measured variable is fed back into the spatial database 144. Since the latest nitrogen measurement is now a "current" assessment of the soil, at least some of the current statements, i.e., the previous nitrogen measurement, are outdated and should be translated into historic statements. Hence, step 164. Again, since the historic statements have been enlarged, the facts are updated 166.

It should be appreciated that the foregoing is only representative of updating the recipe 108 and that still other variations for updating the recipe 108 are embraced within the scope of the present invention. Such other variations include, but are not limited to, updates for the recipe based upon: (i) periodic time intervals such as every 4 months for collecting soil samples or every day as fluctuations in the market price affect the expected earning potential of a yield of harvested crops; (ii) on an as needed basis such as when pertinent information becomes available such as a change in the market price of a bag of fertilizer; (iii) or when events takes place that effect the spatial environment such as atmospheric events like thunderstorms, tornadoes or floods; and any other useful methods.

It should also be appreciated that the present invention advances the present state of for numerous reasons. Some of those reasons include: the advancement of the art by considering the economic impact of every potential action of a farming operation from purchasing fertilizer to managing the equipment schedules of the machinery used to harvest the crops; evaluation of not only the present condition of the field but historic and future conditions as well; - evaluation of the surrounding vicinities such as neighboring and regional spatial environments; ability to combine relevant historical information together with current and predicted information to expand the knowledge of the database by feeding back a result of an actual or predicted event; use of systems such as expert systems and AI routines which leads to trial and error "learning" by the database to grow the database and improve optimization of a recipe; use of flexible methodologies that are not specific to particular crops or spatial 33 environments; and use of systems and methods which only require recipes on an as needed basis to assist in preventing environmental pollution and energy waste.

Example of Fertilization Recipe During 1997, a 135 acre agricultural field near Ashton, Idaho was experimentally prepared for the purposes of generating an economically optimized fertilizer recipe specific to the field. The data was obtained from various and wide ranging sources such as market prices, fertilizer information, adjacent fields during prior growing seasons and sensors in the field. Some data was obtained from spatial sites within the field and some from sources external to the field. The data obtained provided the knowledge for generating statements, both historic and current, and for generating facts of the spatial database. The facts were analyzed against the current statements and a fertilizer recipe specific to the field was devised. The fertilizer was applied to the field by precision farming machinery and closely monitored throughout the growing season. The results, after harvest, demonstrated a cost savings of 39.7% of fertilizer costs per acre. Although a slight decrease of about 2.5 bushels per acre in yield was recorded, the difference in savings on fertilizer more than adequately supported the methodologies used in devising the recipe. Moreover, the decrease in yield can be attributed to the lack of capabilities of conventional precision farming machinery. As described previously, precision farming machinery has a boom for distributing fertilizer which is about 70 feet across. Although highly sophisticated, the precision machinery is still not as advanced as the methodologies taught herein that provide the recipe. Thus, the present invention also advantageously provides the impetus for developing and technologically advancing the state of the farm machinery arts. Eventually, even greater recipe optimization is anticipated.

The present invention may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not restrictive. The scope of the invention is, therefore, indicated by the appended claims rather than 34 by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.

Claims

35WE CLAIM:

1. A method for optimizing a recipe for use in a spatial environment, comprising the steps of: defining a spatial database having a plurality of facts to be associated with said spatial environment; analyzing said facts to determine whether said facts are feasible; and devising said recipe for said spatial environment from said facts that are determined to be feasible, each of said facts that are determined to be feasible are instructions to be performed on said spatial environment to achieve a desired result that is optimized.

2. A method according to claim 1, wherein said step of defining said spatial database further comprises the step of generating a plurality of historic statements from which said facts are generated.

3. A method according to claim 1, wherein said step of defining said spatial database further comprises the step of generating a plurality of current statements, said current statements for analyzing said facts against.

4. A method according to claim 3, wherein said step of analyzing said facts to determine whether said facts are feasible further comprises the step of determining whether said facts can be executed.

5. A method according to claim 4, wherein said step of analyzing said facts to determine whether said facts can be executed comprises a iterative determination for each fact of said facts.

6. A method according to claim 1, wherein said step of analyzing said facts to determine whether said facts are economically feasible comprises a iterative determination for each fact of said facts. 36

7. A method according to claim 1, further comprising the step of updating said recipe.

8. A method according to claim 1, further comprising the step of applying said recipe.

9. A computer-readable medium having computer-executable instructions for performing the steps as recited in claim 1.

10. A method for optimizing a recipe for use in a spatial environment, the spatial environment having aggregate constraints, the method comprising the steps of: defining a spatial database having a plurality of facts to be associated with said spatial database; analyzing said facts to determine whether said facts are feasible in view of the aggregate constraints; and devising said recipe for said spatial environment from said facts that are determined to be feasible, whereby each of said facts that are determined to be feasible are instructions to be performed on said spatial environment to achieve a desired result that is optimized in view of the aggregate constraints.

11. A method according to claim 10, wherein said step of defining said spatial database further comprises: generating a plurality of historic statements from which said facts are generated; and generating a plurality of current statements, wherein said facts are analyzed against said current statements. 37

12. A method according to claim 10, wherein said step of analyzing said facts to determine whether said facts are feasible comprises the step of iteratively determining whether each fact of said facts can be executed.

13. A method according to claim 10, further comprising the step of updating said recipe.

14. A method according to claim 10, further comprising the step of applying said recipe.

15. A computer-readable medium having computer-executable instractions for performing the steps as recited in claim 1.

16. A method for optimizing a recipe for use with an agricultural field, comprising the steps of: generating a plurality of facts to be associated with said agricultural field; generating a plurality of current statements to be associated with said agricultural field, said facts and current statements defining a spatial database for said agricultural field; analyzing said facts and said current statements to determine whether said facts can be executed, each said fact that is able to be executed is a stored fact; and devising said recipe for said agricultural field from each said stored fact that is determined to be feasible.

17. A method according to claim 16, wherein said step of devising said recipe is devised from each said stored fact that is economically feasible.

18. A method according to claim 16, wherein said step of devising said recipe is devised for one of a fertilizer schedule, an irrigation schedule, a 38 herbicide schedule, a pesticide schedule, a seed variety spacing schedule, an agricultural equipment schedule and an information management schedule for said agricultural field.

19. A method according to claim 18, further comprising the step of generating a plurality of historic statements to be associated with said agricultural field, said historic statements being used for generating said facts.

20. A method according to claim 19, wherein said step of generating said historic statements to be associated with said agricultural field comprises the step of using geographical information system routines.

21. A method according to claim 19, wherein said step of generating said facts to be associated with said agricultural field comprises the step of using artificial intelligence routines.

22. A method according to claim 19, further comprising the step of discarding said facts from consideration in devising said recipe when said facts are determined to be in noncompliance.

23. A method according to claim 22, further comprising the step of discarding said stored facts from consideration in devising said recipe when said stored facts are determined to be infeasible.

24. A method according to claim 19, further comprising the step of applying said recipe to said agricultural field.

25. A method according to claim 19, further comprising the step of updating said recipe. 39

26. A method according to claim 25, wherein said step of updating said recipe comprises one of the following steps:

(a) applying said recipe to said spatial environment; measuring a chosen variable; and feeding back said chosen variable to said spatial database as another historic statement, as another fact, or another current statement;

(b) measuring a variable in said spatial environment; translating said current statements into additional historic statements; updating said facts; and

(c) applying said recipe to said spatial environment; measuring a desired result; providing said desired result to said spatial database as another historic statement, as another fact or as another current statement; and translating said current statements into additional historic statements; and updating said facts.

27. A method for optimizing a split recipe for use with an agricultural field, comprising the steps of: generating a first plurality of facts to be associated with said agricultural field; generating a plurality of first current statements to be associated with said agricultural field, said first facts and first current statements defining a spatial database for said agricultural field; analyzing said first facts and said first current statements to determine whether said first facts can be executed, wherein each said first fact that is able to be executed is a stored first fact; devising said split recipe for said agricultural field from each said stored first fact that is determined to be feasible; 40 generating a next plurality of facts to be associated with said agricultural field; generating a plurality of next current statements to be associated with said agricultural field, said next facts and next current statements augmenting the spatial database for said agricultural field; analyzing said next facts and said next current statements to determine whether said next facts can be executed, wherein each said next fact that is able to be executed is a stored next fact; and revising said split recipe for said agricultural field from each said stored next fact that is determined to be feasible.

28. A method according to claim 27, wherein said step of devising said split recipe is devised from each said stored first fact that is economically feasible.

29. A method according to claim 27, wherein said step of devising said split recipe and said step of revising said split recipe is devised and revised for one of a fertilizer schedule, an irrigation schedule, a herbicide schedule, a pesticide schedule, a seed variety spacing schedule, an agricultural equipment schedule and an information management schedule for said agricultural field.

30. A method according to claim 27, further comprising the step of generating a plurality of historic statements to be associated with said agricultural field, said historic statements being used for generating said first facts and said next facts.

31. A method according to claim 30, wherein said first current statements become said historic statements when said next current statements are generated. 41

32. A method according to claim 30, wherein said step of generating said historic statements to be associated with said agricultural field comprises the step of using geographical information system routines.

33. A method according to claim 30, wherein said step of generating said first facts and said next facts to be associated with said agricultural field comprises the step of using artificial intelligence routines.

34. A method according to claim 30, further comprising the step of discarding said first facts and said next facts from consideration when said first facts and said next facts cannot be executed or are not feasible..

35. A method according to claim 27, further comprising the step of applying said split recipe to said agricultural field.

36. A method according to claim 27, further comprising the step of applying said revised split recipe to said agricultural field.

37. A method according to claim 27, further comprising the step of repeating the steps of: generating a next plurality of facts to be associated with said agricultural field; generating a plurality of next current statements to be associated with said agricultural field, said next facts and next current statements augmenting the spatial database for said agricultural field; analyzing said next facts and said next current statements to determine whether said next facts can be executed, each said next fact that is able to be executed is a stored next fact; and revising said split recipe for said agricultural field from each said stored next fact that is determined to be feasible. 42

38. A method according to claim 27, wherein said split recipe is subject to aggregate constraints.

39. A method for optimizing a split recipe for use with an agricultural field, comprising the steps of: generating a plurality of facts to be associated with said agricultural field; generating a plurality of current statements to be associated with said agricultural field, said facts and current statements defining a spatial database for said agricultural field; analyzing said facts and current statements to determine whether said facts can be executed, wherein each said fact that is able to be executed is a stored fact; devising said split recipe for use with a first stage related to the agricultural field; updating said spatial database with at least one temporal statement; generating additional facts to be associated with said agricultural field; and devising said split recipe for use with a next stage related to the agricultural field.

40. A method according to claim 39, wherein said first stage is a period of time.

41. A method according to claim 39, wherein said first stage is determined by crop physiology.

42. A method according to claim 39, wherein said next stage is a period of time. 43

43. A method according to claim 39, wherein said next stage is determined by crop physiology.

44. A method for economically optimizing a fertilizer recipe for use with an agricultural field, comprising the steps of: generating a plurality of historic statements to be associated with said agricultural field, at least some of said historic statements being associated with a historic soil condition of said agricultural field; generating a plurality of facts from said historic statements to be associated with said agricultural field; generating a plurality of current statements to be associated with said agricultural field, at least some of said current statements being associated with a current soil condition of said agricultural field, said historic statements, said facts and said current statements defining a spatial database for said agricultural field; analyzing said facts against said current statements to determine whether said facts can be executed, at least some of said facts being analyzed against said at least some of said current statements associated with said current soil condition to determine whether said at least some of said facts can be used to alter said current soil conditions with a fertilizer blend, each said fact that is able to be executed is a stored fact; and devising said fertilizer recipe for said agricultural field from each said stored fact that is determined to be economically feasible.

45. A method for economically optimizing an irrigation recipe for use with an agricultural field, comprising the steps of: generating a plurality of historic statements to be associated with said agricultural field, at least some of said historic statements being associated with a historic soil condition of said agricultural field; generating a plurality of facts from said historic statements to be associated with said agricultural field; 44 generating a plurality of current statements to be associated with said agricultural field, at least some of said current statements being associated with a current soil condition of said agricultural field, said historic statements, said facts and said current statements defining a spatial database for said agricultural field; analyzing said facts against said current statements to determine whether said facts can be executed, at least some of said facts being analyzed against said at least some of said current statements associated with said current soil condition to determine whether said at least some of said facts can be used to alter said current soil conditions by irrigation, wherein each said fact that is able to be executed is a stored fact; and devising said irrigation recipe for said agricultural field from each said stored fact that is determined to be economically feasible.

46. A computer-readable medium having computer-executable components for providing a user with an optimized recipe for a spatial environment, comprising:

(a) a user interface component for receiving a plurality of statements and facts, said statements and facts being associated with said spatial environment;

(b) a spatial database component for storing said statements and facts, said spatial database being accessible by said user interface component;

(c) an analysis component for comparing said facts with said statements to determine whether said facts are feasible or infeasible; and

(i) for each of said facts determined to be infeasible, discarding said fact from consideration in said recipe; and

(ii) for each of said facts determined to be feasible, including said facts as part of said recipe, said recipe being a set of instructions to be performed for achieving a desired result for said spatial environment that is optimized. 45

47. The computer-readable medium according to claim 46, wherein said user interface component is further able to receive said statements as either an historic statement or a cunent statement, said spatial database component being capable of separately storing said historic and current statements.

48. The computer-readable medium according to claim 47, further comprising an updating component for interpreting when said spatial database is enlarged to translate said current statements into additional historic statements or additional facts.

49. The computer-readable medium according to claim 48, wherein said updating component is further able to update said facts when said current statements are translated into said additional historic statements.

50. The computer-readable medium according to claim 46, wherein said analysis component is further able to determine whether said facts are able to be executed, if said facts are able to be executed said facts are maintained as stored facts, if said facts are not able to be executed said facts are discarded from consideration of said recipe.

51. The computer-readable medium according to claim 50, wherein said analysis component is further able to reiteratively determine whether each of said facts are feasible and to iteratively determine whether each of said stored facts is able to be executed.

52. The computer-readable medium according to claim 46, wherein said analysis component is able to determine which of said facts that are feasible should be executed to achieve the optimized recipe when said spatial environment is subject to aggregate constraints. 46

53. A method for optimizing a recipe for multiple schedules for use with an agricultural field, comprising the steps of: generating a plurality of facts to be associated with said agricultural field; generating a plurality of current statements to be associated with said agricultural field, said facts and current statements defining a spatial database for said agricultural field; analyzing said facts and said current statements to determine whether said facts can be executed, each said fact that is able to be executed is a stored fact; and devising each said recipe for each said schedule for said agricultural field from each stored fact that is determined to be feasible.

54. A method according to claim 53, wherein said schedules are one of a fertilizer schedule, an irrigation schedule, a herbicide schedule, a pesticide schedule, a seed variety spacing schedule, an agricultural equipment and an information management schedule.

55. A method according to claim 53, wherein said step of devising each said recipe for each said schedule is devised from each stored fact that is economically feasible.

56. A method according to claim 53, further comprising the step of generating a plurality of historic statements to be associated with said agricultural field, said historic statements being used for generating said facts.

57. A method according to claim 53, further comprising the step of applying said recipe to said agricultural field.

58. A method according to claim 53, further comprising the step of updating said recipe. 47

59. A method according to claim 53, wherein the agricultural field is subject of aggregate constraints.