Tải bản đầy đủ - 0 (trang)
II. Overview of the Basic Components of Precison Farming

II. Overview of the Basic Components of Precison Farming

Tải bản đầy đủ - 0trang



cy of these site-specific practices in order to assess value on and off the farm. Thus,

precision agriculture is technology enabled, information based, and decision focused (Pierce, 1997a).



While the concept of matching inputs to site-specific conditions is not new, as

just discussed, there is little doubt that important advances in technology continue to enable precision agriculture. The enabling technologies of precision agriculture can be grouped into five major categories: computers, global position system

(GPS), geographic information systems (GIS), sensors, and application control.

Few of the enabling technologies were developed specifically for agriculture and

their origins date back more than 20 years, as illustrated in the time chart in Figure 1. It is the integration of these technologies that has enabled farmers and their

service providers to do things not previously possible, at levels of detail never

before obtainable, and, when done correctly, at levels of quality never before

achieved (Fortin and Pierce, 1998).

1. Computers

Many technologies support precision agriculture, but none is more important

than computers in making precision agriculture possible. Also, it is not computers

alone that are important but their ability to communicate that is so powerful for

agriculture.As Taylor and Wacker (1997) suggest, it is the fusion of computers and

communication that gave birth to connectivity, and it is connectivity that is driving

the access of everyone to everyone, everything to everything, and everything to

everyone. This electronic linkage and communication define the age of access

(Taylor and Wacker, 1997). It is this notion that may have prompted the NRC

(1997) to define precision agriculture in terms of a management strategy that uses

information technologies for decision making.

Precision agriculture requires the acquisition, management, analysis, and output of large amounts of spatial and temporal data. Mobile computing systems were

needed to function on the go in farming operations because desktop systems in the

farm office were not sufficient. These mobile systems needed microprocessorsthat

could operate at speeds of millions of instructions per second (MIPS), had expansive memory, and could store massive amounts of data. The first microchip created by Intel in 1971 (Intel 4044 processor) contained a mere 2300 transistors and

performed about 60,000instructions per second. Since 1971, the number of transistors per chip has doubled every 18 months (Fig. 2) affirming Gordon Moore’s

observation in 1965 that a doubling of transistor density on a manufactured die

was occurring every year, a concept referred to as “Moore’s law” (Moore, 1997).




Aerial photography emerges; pictures taken from balloons

First image sensors incorporated in satellites; low-resolution black and white TV

First commercial CIS

First chlorophyll sensor (Benedict and Swidler, 1961)

First multispectral photography done from space Apollo 9 manned mission

Baumgardener et al. (1970) related soil organic matter to multispectral data

Intel 4040 processor

Launch of Earth Resources Technology Satellite-1 later renamed Landsat; permitted

continuous coverage of most of the earth's surface

Soil organic matter sensor (Page, 1974)

Apple computer commercialized (http://www.apple.com)


























Launch of first NAVSTAR GPS satellite

First IBM PC

Intel 80286 processor

Launch of Landsat "Thematic

Mapper ("h4) added

The Jet Propulsion Lab produces hyperspectral sensors for use from a high-altitude

aircraft platforms known as AIS (Airborne Imaging Spectrometer)

GPS available for civilian use

Ortlip patent issued to SoilTEQ

Launch of Landsat 5

286 Intel processor

Grain flow monitoring on combines @e Baerdwmeker et al., 1985)

French launch an operational series of earth-observingsatellites called SPOT (SysPme

Probatoire d'observation de la Terre); first offering of multispectral data to world

users on a commercial basis

The Jet Propulsion Lab produces a second hyperspectral sensor known as AVIRIS

(Airborne VisibldnfraRed Imaging Spectrometer)

Yield mapping in Texas (Bae et al.. 1987)

India launches earth resources satellite (IRS-IA) that gathers data in the visible and

near IR with the Linear Imaging Self-scanning sensor (LISS)

Intel 40486 processor

Canadian Radarsat, ERS-I, and ERS-2 managed by the European Space Agency A

class of satellite remote sensors using radar systems

Japan launches JERS- 1 and JERS-2 that include both optical and radar sensors

Selective availability (SA) imposed on GPS signal

First symposium on site-specific crop production (ASAE, 1991)

Commercial yield monitors appear in the United States

First international conference on soil specific crop management (Robert et al., 1993)

Pentium processor

Full constellation of 24 GPS satellites in NAVSTAR system complete

Earth System Science Pathfinder launched by NASA

Pentium I1 processor

India launches the latest in the series. IRS-ID, on September 29, 1997

First European conference on precision agriculture (BIOS, 1997)

Board of Agriculture, National Research Council report on precision agriculture


Figure 1 Historical developments in the technologies that enabled precision agriculture.




Figure 2 Illustration of Moore’s law showing the doubling of computer speed and capacity every

year [Source:Intel Corporation (www.intel.com)].

As Fig. 2 indicates, Moore’s law is expected to hold until 2017 (according to

Moore) and appears to hold for memory and storage capacity. Data storage capacity will need to increase rapidly as sensor technology and digital geospatial data

become increasingly available to agriculture. Moore notes,

By the Year 2012, Intel should have the ability to integrate 1 billion transistors

onto a production die that will be operating at 10 GHz. This could result in a

performance of 100,OOOMIPS, the same increase over the currently cutting edge

Pentium I1 processor as the Pentium 11processor was to the 386! We see no fundamental barriers in our path to Micro 2012, and it’s not until the Year 2017 that

we see the physical limitations of wafer fabrication technology being reached.

We can expect, therefore, that computers will drive significant technological development to enable precision agriculture for the foreseeable future. The extent to

which agriculture can utilize computer technology is important to the success of

agriculture in general (Holt, 1985; Ortmann et al., 1994). However, the agricultural sector is lagging in the adoption of computer technologies on the farm relative to other business sectors. According to the 1997 annual survey of the U.S. Department of Agriculture (USDA) National Agricultural Statistics Service (NASS,

1997), of 2,053,800 farms in the United States, only 38% had computer access,

31% owned or leased a computer, and 13% had Internet access. Part of this computer lag in agriculture is due to the lack of access or connectivity in rural areas,

lack of training, and little perceived utility in available software (Nowak, 1997;



Peterson and Beck, 1997).In any event, farmers will have to become as comfortable working with computers and their data as they are working with their farm

machinery (Klein, personal communication, 1997).

While it appears that computer hardware will be more than adequate for precision agriculture, the same cannot be said for the software. Advances in software

logically lag behind the hardware technology. However, software for precision

agriculture has been more an experience than an application. Berry (1999, in discussing the human factor in GIS, describes experience as “what you get when you

don’t get what you want.” Computer software in precision agriculture has become

better with time, but precision agriculture is loaded with Berry’s type of experience. Software will be adequate for precision agriculture when it becomes, as

Berry (1 995) suggests, second nature to the user for assessing information and

translating it into knowledge. For precision agriculture, the knowledge needed is

that for managing variability on the farm, knowledge that is requisite for decision

making. Computers and salient, usable software are going to play a critical role in

the emergence of a precision agriculture system in the near future.

2. Geographic Information Systems

Formally, GIS is an organized collection of computer hardware, software, geographic data, and personnel designed to efficiently capture, store, update, manipulate, analyze, and display all forms of geographicallyreferenced information [Environmental Systems Research Institute (ESRI), 19971. The GIS concept dates

back to the 1960s when computersbecame available for use in spatial analysis and

quantitative thematic mapping (Burrough, 1986).The science of GIS has evolved

since the 1960s to include data management and modeling, enabling a shift from

mapping to spatial reasoning (Berry, 1993, 1995). The ability to perform spatial

operations on the data distinguishes a true GIS from the many software programs

that do thematic mapping and database management. During the past few years,

mapping software programs have been adding spatial operations, workstation GIS

software programs have spawned microcomputer versions with more limited GIS

capabilities to fit desktop computer technologies, and new microcomputer-based

GIS systems have emerged. There are many different mapping and GIS software

programs that offer different GIS features. None, however, have captured the market for application in precision agriculture.

Because precision agriculture is concerned with spatial and temporal variability and because it is information based and decision focused (Pierce, 1997a), it is

the spatial analysis capabilitiesof GIS that enable precision agriculture. This statement is true because the value of precision agriculture is derived only when resulting information is turned into a management decision that increases profitability, benefits the environment, or provides some other value to the farm. AGIS,

in the full sense of its formal definition given previously, is key to extracting val-



ue from information on variability. Clark and McGucken (1996) refer to GIS as

the brain of a precision farming system. However, available GIS software packages are complex and difficult to learn for nonspecialists. Some GIS lack data management and spatial analysis tools needed to understand the variability observed

on the farm and needed to derive site-specific management recommendations.

More functional, easy to use interfaces are needed in order to fully utilize this technology in production agriculture (Berry, 1995; NRC, 1997).Computer simulation

modeling can help derive the needed understanding of variability (Sadler and Russell, 1997; Verhagen and Bouma, 1997) and linking GIS to models (Goodchild et

al., 1993) will be important to precision agriculture.

3. Global Positioning Systems

Location control (Schueller, 1992) is essential to precision agriculture for assessing spatial variability and for site-specific application control (Auernhammer,

1994; Tyler et al., 1997). In the early days of precision agriculture, relative position within a field was determined by dead reckoning. This was a simple method

in which position was measured relative to a known point determined by measuring distance using radar, ultrasound, and wheel shaft counters. Direction was determined either by using a steering-angle sensor or a gyroscope from the known

point or by direction only if a field had linearized tramlines of known fixed location (Auernhammer and Muhr, 1991). Triangulation methods, in which position is

determined relative to two or more known locations using, for example, radio signals transmitted from reference stations to mobile units (Palmer, 1991, 1995; Scorer, 1991), improved position accuracy to as low as 15 cm (95%probability) but

such systems were time-consuming and expensive. By the early 1990s. however,

the GPS known as NAVSTAR (NAVigationSystem with Time And Ranging) was

becoming available for general civilian use including agriculture.This system was

based on 18 satellites that were in orbit by early 1990 (Hoffmann-Wellenhofet al.,

1994; Kaplan, 1996; Kennedy, 1996; Leick, 1995; NAPA, 1995). The United

States NAVSTAR GPS system consists of a constellation of 24 satellites, including 3 spares. The first satellite was launched in 1978 but it was not until the Soviet downing of a Korean airliner in 1983 that the decision was made to make GPS

available for civilian use [National Academy of Public Administration (NAPA),

19951. The NAVSTAR GPS system was fully deployed by 1994 and declared

fully operational in 1995. The Russians also deployed a GPS system called

GLONASS (Global Navigation Satellite System) consisting of 24 satellites completed in 1995. Although there are differences in time standards and coordinate

systems between GLONASS and NAVSTAR, higher end GPS receivers currently

available accommodate the combined use of both GPS systems resulting in increased reliability and accuracy. Although the Russian GLONASS policy called

for ensured availability for 15 years, no charge on a constant global basis, and no

selective availability, the system was degraded to only 14 or 15 active spacecraft



during the fall of 1997 (Perry, 1998). Therefore, changes in GPS technology are

to be expected.

The GPS technology enables precision agriculture because all phases of precision agriculture require positioning information. GPS is able to provide the positioning in a practical and efficient manner for a few thousand dollars ( v l e r et al.,

1997). Expensive, high-precision differential GPS (DGPS) systems are available

that achieve centimeter accuracies (Lange, 1996), allow for automated machinery

guidance (O’Conner et al., 1996; 51er et al., 1997) and kinematic mapping of

topography (Clark, 1996), and are useful in the creation of digital elevation models needed for terrain analysis (Bell et al., 1995; Moore et al., 1993). While the

GPS signal is ubiquitous, there have been problems in making available GPS at

the needed precision for agriculture (Saunders et al., 1996). The U.S.Department

of Defense implemented selective availability (SA) on March 25, 1990, which limited accuracy of GPS to civilians from about 8-10 m without SA to about 100 m

with SA. This was done by varying the reported precise time of clocks on board

the satellites and by providing incorrect orbital positioning data (NAPA, 1995).

The SA has been overcome by the use of differential corrections transmitted to

GPS receivers (rovers) from GPS receivers at known fixed locations (base). DGPS

involves the transmission of a differential correction, that is, the difference between actual and predicted position at the base GPS receiver, to rover GPS receivers, which then apply the corrections to received GPS signals to solve for a

more accurate position (Qler et al., 1997). There are four general ways of providing a differential correction: a private local GPS base receiver with a radio modem to transmit to a mobile receiver, a commercial GPS base station at which differential corrections are transmitted on FM subcarrier frequencies, a public GPS

base station at which differential corrections are transmitted on AM frequencies

from radio beacons with up to a 250-mile radius [U.S. Coast Guard (USCG) beacon system), and a wide area differential GPS (WADGPS) network in which differential corrections from a network of base stations are used by the roving GPS

receiver to correct its position (vier et al., 1997). In all cases, DGPS requires additional receivers and antennas and is fee based for commercial correction

providers. A differentialcorrection is desirable even without SA because it is needed to achieve the accuracies needed in some aspects of precision agriculture, including navigation and guidance. Currently, only WADGPS provides national coverage, whereas all others are dependent on whether the rover is close enough to a

base station to receive the signal consistently. However, this is changing because

FM providers are planning to offer national coverage in the near future and there

are plans for completion of the USCG beacon system nationally (Divis, 1998).

There is currently a debate as to whether the public sector should provide a national DGPS (NDGPS) to agriculture (NAPA, 1995;Pointon, 1997). Other sectors

of the U.S. economy also need a national NDGPS, so the discussion of who benefits from a publicly supported NDGPS should not be focused on agriculture alone.

The Office of Management and Budget did not support expansion of the USCG



radio beacon system for public NDGPS in Ey98. However, some believe that a

government-provided NDGPS system is so important to critical activities that it is

best for the government to provide it (Divis, 1998). Certainly, precision agriculture needs DGPS and will require increased position accuracy as new technologies for navigation and guidance require higher precision, which may require

DGPS accuracies not available from a government NDGPS. The prophecy of

Auernhammer and Muhr (1991; p. 395) that “their use will also be without costs

in the future” will probably never be realized because DGPS is big business.

Regardless of who provides all aspects of DGPS, farmers and their service

providers need reliable DGPS to achieve the desired positioning for precision

farming operations. Farmers still experience interruptions and interferences in the

GPS and/or differential correction signals, creating gaps in data collection or loss

of application or guidance control. In activities at higher speeds, such as aerial applications (Kirk and Tom, 1996), time delays in differential corrections may limit

positional accuracy in kinematic mode (NRC, 1997). The specified availability

(four satellites in view at any location) of the NAVSTAR system is 99.85%, with

a reliability (system is in service when it needs to be) of 99.97% (NAPA, 1995).

However, the suitability of the satellite geometry for calculating a solution, referred to as dilution of precision (DOP), is a problem in farming in which natural

or man-made structures obstruct the receivers’ view of some satellites or interfere

with differential correction reception. There are also geographic locations at which

DOP has been inadequate for needed location precision at certain times during the

day. Additionally,some GPS receivers are susceptible to unwanted interfering signals from a variety of sources, including farm machinery, making the receiver useless in navigation or positioning. Some interferences can be overcome in the design of the GPS receiver.

Regardless of problems, DGPS has greatly enabled precision agriculture. Of

great importance for precision agriculture, particularly for guidance and for digital elevation modeling, position accuracies at the centimeter level are possible in

DGPS receivers that use carrier phase in combination with DGPS (Lange, 1996;

Tyler er al., 1997).Accurate guidance and navigation systems will allow for farming operations not currently in use, including field operations at night when wind

speeds are low and more suitable for spraying and the use of night tillage to reduce the light-induced germination of certain weeds (Hartmann and Nezadal,

1990). DGPS technology changes continually and can be followed on the internet

(e.g., Peter Dana’s web site hrtp://wwwhost.cc.utexus.edu/Stp/pub/grg/gcrafr/

notes/gps/gps.html or www.gpsworld.com).

4. Sensors

Sensors are devices that transmit an impulse in response to a physical stimulus

such as heat, light, magnetism, motion, pressure, and sound. With computers to



record the sensor impulse, a GPS to measure position, and a GIS to map and analyze the sensor data, any sensor output can be mapped at very fine scales. Sensor

technology currently lags behind other enabling technologies (Sudduth et al.,

1997) and the availability of sensors has been cited as the most critical factor preventing the wider implementation of precision agriculture (Stafford, 1996b). Sensors are critical to success in the development of a precision agricultural system

for three important reasons: Sensors have fixed costs, sensors can sample at very

small scales of space and time, and sensors facilitate repeated measures. This

means that the cost per sample is determined by the extent of sensor use, sample

intensity is determined by the capability of the sensor and not the cost or difficulty in sampling associated with traditional physical sampling schemes, and sampling frequency is determined by accessibility of the target and not costs.

The value of sensors and their potential for the future of precision agriculture are

illustrated by yield monitoring.Yield monitoring systems, which use sensorsto measure crop flow, allow the creation of yield maps with detail not practical with other

measurementtechniques (Pierce er al., 1997).Yield mapping technology may be the

major factor responsible for the growing interest in precision agriculture observed

since its commercial introduction in 1992 (Stafford, 1996b). Prior to 1992, the focus was on VRT, which would not in itself have sustained precision agriculture.

Yield mapping bolstered precision agriculture and is currently the major precision

agriculture technology in U.S. agriculture. However, the promise of sensing technologies may make yield mapping technology unnecessary in the future if high-resolution remote sensing of the growing crop leads to quantitative prediction of crop

yield prior to harvest. Yield mapping will serve to validate sensor-based predictive

technology, but once operational, yield monitors may not be needed. The use of remote sensing to forecastcrop yields is in use worldwide, and forecastingoffers farmers the ability to market their crops prior to harvest when prices are more favorable.

Sensors are very desirable for use in precision agriculture. Every effort should

be made to promote the application and adaptation of sensors developed in other

industrial sectors, especially the space and defense industries, as well as to promote the development of new sensor technologies for use in assessing and managing variability in soils, plants, pests, and machinery. Sensors can be contact or

remote, ground based or space based, and direct or indirect. Sensors have been developed to measure machinery, soil, plants, pests, atmosphericproperties, and water by sensing motion, sound, pressure, strain, heat, light, and magnetism and relating these to properties such as reflectance, resistance, absorbance, capacitance,

and conductance. Sensors are needed in precision agriculture because such a system requires the collection,coordination, and analysis of massive quantities of data

(Sudduth et al., 1997),some for strategic surveys and inventories and some for use

in real-time applications.

Remote sensing involves the detection and measurement of photons of differing energies emanating from distant materials. These photons may be identified



and categorized by classhype, substance, and spatial distribution, with most designed to monitor reflected radiation (Frazier et al., 1997). Satellite remote sensing dates back to the first aerial photographs taken from balloons in the 1840s.The

first satellite imagery was obtained from TV cameras mounted in satellites in the

early 1960s. Since the U.S.Landsat program launched the first observation satellite in 1972, earth observation has increased and currently India, France, Russia,

Japan, and the European Space Agency also operate earth observation satellites

(Figure 1). Many companies now offer commercial products to agriculture from

images obtained by these satellites or enhanced digital products derived from

them. Remote-sensing satellites collect image data actively by sending a known

signal from the satellite to the earth and measuring the portion of the signal that is

returned. Passive data collection occurs by measuring the incoming energy from

the sun reflected by an object or heat energy emanated from an object. The electromagnetic energy emanated from an object varies in wavelengths as determined

by the object’s physical and chemical structure. Different images of an object can

be constructed by combining different wavelengths, creating images far more revealing than images obtained from visible light alone. Remote-sensing systems

vary in spatial resolution (meters to kilometers), spectral coverage (portion of the

light spectrum covered), and temporal frequency (days to months). Different applications in agriculture will require different spatial resolutions, spectral coverages, or temporal frequencies. NASA (1998) provides an online tutorial on remote

sensing and its applications. Moran et al. (1997) provide a comprehensive review

of image based remote sensing for precision agriculture.

Remote sensing holds great promise for precision agriculture because of its potential for monitoring spatial variability over time at high resolution (Hatfield and

Pinter, 1993; Moran e? al., 1997; Stevens, 1993). For example, monitoring of a

growing crop using remote sensing is critical because yield maps document yield

variability but do not provide information on the cause of observed variability.

However, the promise of remote sensing for agriculture has not been realized for

many reasons, including costs, timeliness, and availability (Frazier et al., 1997;

Stafford, 1996b).

5. Application Control

Control is that portion of an automated system in which sensed information is

used to influence the system’s state in order to meet an objective (Stone, 1991).

For precision agriculture, control must be achieved in space and time for varying

single or multiple inputs at different rates, at varying soil depths, and in a uniform

and location-specific manner within fields. Because it is a required component,

control technology has been a strength of precision agriculture since its inception

and the state of application control was recently reviewed by Anderson and Humburg (1 997). Simply stated, if the needed accuracy cannot be achieved at the point



of application of inputs, then precision agriculture cannot be successful (Anderson and Humburg, 1997; Schueller, 1992; Stafford, 1996b). Application control

completes the precision agriculture loop.

Control systems are currently available at varying degrees of precision for variable seed and metering granular fertilizers and pesticides, changing varieties on

the go, anhydrous N application, sprayers, imgation, manure application, and various tillage implements (Anderson and Humburg, 1997; ASAE, 1991; Robert el

al., 1993, 1995, 1996). The first patented technology for variable rate application

of fertilizers was the Ortlip patent awarded in 1985 to Soil-TEQ, Inc. (now owned

by AgChem), although the earliest references to precision application of fertilizers appear to be Luellen ( 1985) and Elliot ( 1987).

All issues relating to the accuracy of application equipment are important to precision agriculture but not all accuracy issues are unique to precision agriculture

(Anderson and Humburg, 1997). General sources of variability in application of

inputs include driving precision, uniformity of distribution, topography, field surface conditions, wind conditions, and metering efficiency. Specific to precision

agriculture are the transition time for changes in rate or product and positioning or

location control and those aspects of application in which changing rates or products affect variability in performance. A high precision, absolute reliable DGPS

will offer the positioning precision required for various tasks in precision agriculture. Some argue for a backup system, such as dead reckoning, to avoid loss of

control if DGPS fails (Auernhammer and Muhr,1991). While very high position

accuracy is available using DGPS, currently the major consideration is cost. Human driving precision has an expected coefficient of variation of 10% for moderate skill levels (Chaplin et al., 1995) but should be greatly improved with DGPSbased guidance systems, depending on the accuracy of the DGPS system in use

(51er el al., 1997). O'Conner et al. (1996) report the use of a carrier phase differential GPS for automatic vehicle control to achieve a vehicle position accuracy

within a few centimeters and heading to within 0.1".

The issue of transition time is illustrated by the V-shaped spray pattern resulting from a transport delay incurred between the injection point and the nozzle discharge for a simple chemical injection system (Steward, 1994, as cited by Anderson and Humburg, 1997). Transition times of 3-9 s were reported by Bahri et al.

(1996) when changing seeding rates in grain drills, with transition time depending

on the magnitude of the application rate change. Their data indicated that small

rate changes in seeding rate of 10 kg ha-' did not provide a real rate change, illustrating potential step size rate limitations for some inputs. A transition phase

may limit the spatial resolution of variable rate application of inputs if the target

area is small (Stafford, 1996b)or may cause applicationerrors if the transition time

is greater than the time between detection of the need for change and the equip

ment arrival at the detected position as would occur in real-time application.

Current equipment may not be suitable for precision agriculture. Bashford



(1993) and Bashford et al. (1996) report outlet CVs for grain drills ranging from

12 to 22.5%for wheat (Triticumaestivuum L.) and from 16 to 42%for soybean

[Glycine 1 7 1 ~(L). Merr.]. They suggest that external fluted metering devices are

not suitable for precision agriculture. Bahri et al. (1996) measured down-the-row

CVs ranging from 10 to 19%.In general, variation in grain drills has been considered acceptable if the variability in grain and fertilizer delivery among row units

is below a CV of 15%(Prairie Agricultural Machinery Institute, 1987). This variance, however, may exceed the desired accuracy in precision seeding systems.

Centrifugal fertilizer spreaders are known to have high sensitivity of the spread

pattern to flow rate variations and efforts are under way to design centrifugal

spreads for precision agriculture (Olieslagers et al., 1996; Kaplan and Chaplin,

1996). Rate changes also affect nozzle performance relative to drop size and flow

rate for the given nozzle design (Anderson and Humburg, 1997) and these issues

are currently being addressed (Giles et al., 1996; Nuspl et al., 1996).

The major issues for precision application of inputs remain transition time for

changes in product or rate, uniformity of application, and rate increment control.

There are other issues affecting the availability or performance of application control, including the development of standards for communication and connectivity

among manufacturers, a topic being addressed by many organizations including

an association of industry and the public sector called the Ag Electronics Association, ASAE, and the International Standards Organization. There are laws regulating fertilizer quality that in some states (e.g., Arkansas and Michigan) limit the

blending of fertilizers on the go because of the need for a guaranteed chemical

analysis, and such laws will have to be properly addressed. There are issues related to equipment wear (Ballal et al., 1996) and to weather conditions at the time of

application. Kirk and Tom (1996) report that up to 13%of the variability in their

tests for spray aircraft was due to wind conditions. Heterogeneity in the composition of some materials affects the flow or spreading properties (e.g., manures; Ess

et al., 1996).Topography and field surface conditions also affect accuracy, in part

due to their effect on flow of materials in the hoppers or tanks.

Application control, including navigation and guidance, has been enabling precision agriculture since its inception and continues to improve. Farmers and their

service providers have the capability to apply very precise applications of inputs

site specifically. Application control technology will continue to improve and support the needs of precision agriculture. What is needed is knowledge of what inputs are required where and when.


The basic steps in precision agriculture are assessing variation, managing variation, and evaluation. While the enabling technologies facilitate precision agri-

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

II. Overview of the Basic Components of Precison Farming

Tải bản đầy đủ ngay(0 tr)