Nof S.Y. Springer Handbook of Automation

Подождите немного. Документ загружается.

Automation in Agriculture 63.4 Animal Automation Systems 1115

Fig. 63.18 Aquaculture closed system with a feeding sta-

tion. Feed quantities are calculated on a daily basis to each

ﬁsh tank according to ﬁsh weight, water temperature, and

growth rate

devices [63.143]. In addition feeding rhythms affect

feed conversion rates and proximal composition of ﬁsh

ﬂesh. Automated feeding systems (see, e.g., Fig. 63.18)

include timer-controlled feeders [63.144],demand feed-

ers, and automated data-acquisition systems to assess

ﬁsh feeding rhythm, and acoustic, photoelectric sensors

to detect the turbidity of the efﬂuent. Hydroacoustic

sensors and machine-vision systems have been used to

detect left-over pellets.

Future research should be directed towards en-

gineering environmental monitoring and controlling

recirculated systems, and the development of sustain-

able automated systems. Considering that sustainable

development is probably the major challenge faced by

aquaculture [63.145, 146], one should consider sus-

tainability, which can be considered in three main

categories: environmental, economical, and sociolog-

ical [63.147]. Another perspective of sustainable de-

velopment relates to resource utilization and external

effects that are described by various indicators (mainly

in physical terms [63.136]). This should include on-

line reporting of system failures and automation of

the ﬁnal harvesting and grading process [63.148],

thereby improving food safety and maintaining product

quality.

63.4.3 Poultry

Poultry house controllers include sensors for inter-

nal and external temperature measurement, moisture,

static pressure, feed lines, water consumption, and gas

and vent box status [63.149, 150]. Additional automa-

tion equipment includes feed consumption monitoring

equipment, bird weight scales, feed bin load sensors,

gas meters, and water meters.

Physiological signals are important for health moni-

toring and behavior analysis. Several systems have been

developed, including an implanted radiotelemetry sys-

tem for remote monitoring of heart rate and deep body

temperature, and multispectral image analysis for real-

time disease detection [63.151]. An automated growth

and nutrition control system has been developed for

broiler production using an online parameter estima-

tion procedure to model the dynamic growth of broiler

chickens as a response to feed supply [63.152]. Im-

age based bird behavior analysis can be can be used

to develop time proﬁles of bird activity (movement,

response to ventilation, huddling, etc.) as well as to

compare activity levels in different portions of the

house. Time proﬁles of bird activity can contribute to

improved feeder and water design, and enhanced distri-

bution of ventilation air to provide more uniform bird

comfort [63.149,150,153].

Several mechanical poultry catching systems

[63.154] have led to improvements in bird welfare

in addition to manual labor reduction. Systems in-

clude [63.154]: rubber paddles that rotate onto the birds

from above and then push the birds onto a conveyor belt

which carries them back to a loading platform where

they are deposited into crates, a hydraulic drive sys-

tem that advances along the poultry house and picks up

the birds with soft rubber-ﬁngered cylinders that gen-

tly lift them onto a conveyor that transfers the birds

to a caging system, and the Anglia Autoﬂow (Norfolk,

UK) batch-mode catcher that shuttles birds from collec-

tion to a separate packing unit.

63.4.4 Sheep and Swine

Robot shearing operations have been developed and

commercially applied in Australia [63.155]. The sheep

is constrained with straps on a movable platform.

Hydraulically position clippers using force feedback

control the actual shearing of the wool. Path compu-

tations are continuously updated during the shearing

process.

Several feeding systems exist in sow farming: com-

mercial electronic feeding systems that feed one at

a time by enclosing each sow as it eats, electronic

sow feeding systems in loose housing environments

that limit the feed ration [63.156], and a computer-

controlled system that allows sows to feed from one

of two feed formulations to meet their nutritional re-

Part G 63.4

1116 Part G Infrastructure and Service Automation

quirements while satisfying their need for satiety by

using bulk ingredients providing automatic body weight

and average daily weight gain [63.157]. An important

indicator of animal growth and health is the animals’

weight, in addition to its importance in determining

readiness for market. Weighing has been accomplished

using walk-through weighing based on mechanical

scales and imaging systems [63.158]. Physiological

variables measurements include body shape and size

using image analysis [63.159]. Ultrasonic probes have

been applied to measure back fat for monitoring an-

imal growth and feeding regimes. A robotic system

capable of holding a sensor and placing it on the pig

while it is located in the feeding stall has been devel-

oped [63.160].

Real-time behavior and control of swine ther-

mal comfort has been achieved using imaging sys-

tems [63.161]. Planning individual showering systems

for pregnant sows to prevent heat stress [63.162]as

been used in automatic shower cages to prevent waste

water and improve efﬁciency. Automatic cleaning sys-

tems to reduce infections risks between batches of pigs

has been used based on an intelligent sensor for robotic

cleaning [63.163].

Recent environmental policies limiting the amount

of nitrogen and phosphorus that can be applied in the

ﬁeld have led to the development of online analysis

of pig manure systems, including mobile spectroscopy

instruments in the visible and near-infrared wave-

bands [63.164].

63.5 Fruit Production Operations

Fruit production automated systems deal with all stages

of production: growing (automated sprayers, weeders),

harvesting, and post harvest operation (grading, sort-

ing).

63.5.1 Orchard Automation Systems

Fruit production operations in orchards such as pruning,

thinning, harvesting, spraying, and weeding have been

mechanized andautomated. Even when automation sys-

tems have been developed for the same variety of

fruit tree, their components differ substantially because

plant training systems, cultivation methods, climate

conditions, labor conditions, and other conditions and

situations differ from country to country. This section

describes functions, mechanisms, and important obser-

vations of automation systems in orchards.

Fruit Harvesting Robots in Orchards

Several types of shakers are working in orange fruit

orchards: trunk shake and catch, mono-boom trunk

shake, canopy shake and catch, continuous canopy

shake, and others. These shakers are used due to labor

shortage, but the harvested fruits are only for pro-

cessing into juice; they cannot be consumed in the

fresh market because of unavoidable damage. Several

types of orange harvesting robots that have manipu-

lators with picking end-effectors and machine-vision

systems have been reported in the USA, Japan, and

European countries [63.165–169]. Figure 63.19 shows

an articulated manipulator with three degrees of free-

dom (DOFs) mounted on the base attached to the boom.

It was developed by Kubota Co., Ltd., Japan. The

advantage of the articulated manipulator is its com-

pact size when folded up in a narrow space between

trees. Figure 63.20 shows a prismatic arm with three

DOFs driven by hydraulic power. Citrus trees have large

canopies and many branches, twigs, and leaves. Since

these can often be obstacles for fruit harvesting, re-

search on robots with more degrees of freedom has

also been reported [63.170, 171]. Color cameras were

often used as sensing systems to detect fruit because

citrus fruit have orange colors. Fruit locations are cal-

culated by use of stereo vision, differential object size,

vision servoing, ultrasonic sensors or a combination of

them [63.172–179]. Their end-effectors have the func-

tion of rotating semicircular cutters so that they can cut

peduncles in various directions.

Fig. 63.19 Orange harvesting robot (Kubota Co., Ltd.)

Part G 63.5

Automation in Agriculture 63.5 Fruit Production Operations 1117

Fig. 63.20 Orange harvesting robot (University of Florida)

Grape [63.180, 181], apple [63.182, 183], melon

[63.184], watermelon [63.185], and other fruit harvest-

ing robots [63.186] have also been studied. The basic

mechanisms of the manipulators depend on fruit tree

canopy size and shape. Grapevines in many of Euro-

pean and American countries are grown in crop rows,

but those in Asian countries are grown on a trellis

training system due to different climate conditions.

Melons and watermelons are grown on the ground.

There is research on fence-style training systems for

orange trees [63.187] for higher-quality products. This

approach of changing the training system takes a horti-

cultural approach to accomplish a higher rate of success

for harvesting robots. Since the harvesting operation is

usually conducted once in a year in orchards, during

a short period, a robot which can only harvest fruits

is not economical. Therefore, other operations such as

thinning, bagging, and spraying are needed for the or-

chard robot. Some of these functions are accomplished

by replacing end-effectors and software [63.188].

Automation of Spraying

and Weeding Operations

Control of disease, insect pests, and weeds is an es-

sential operation to gain a stable high yield of crops

and high-quality products. This operation includes bi-

ological, physical, and chemical methods. Chemical

spraying is widely used in agricultural production en-

vironments. Today, technologies with high accuracy

to spray only the necessary parts of the plant using

a minimum amount of chemicals are required to protect

workers as well as the environment.

A nozzle-positioning system for a precision sprayer

was studied with a robust crop position detection sys-

Manipulator

Guide rollers

Ridge

End-effector

Fig. 63.21 A multioperation robot

tem at Tohoku Experimental Station, Japan [63.189]

under varying ﬁeld light conditions in rice crop ﬁelds.

The data from a vision sensor was transmitted to a her-

bicide applicator that is made up of a microcontroller,

with slidable arms coupled with spray nozzles. Nozzles

were driven to the optimal positions. The system was

tested to evaluate its performance. It had high enough

accuracy for use in Japanese rice crop ﬁelds. A ﬂuid-

handling system to allow on-demand chemical injection

was developed for a machine-vision-controlled sprayer.

The system was able to provide a wide range of ﬂow

rates of chemical solution [63.190].

Wiedemann et al. [63.191] developed a spray boom

that could sense mesquite plants. Sprayers were at-

tached to tractors and all-terrain vehicles. Controllers

were designed to send ﬁxed-duration voltage pulses to

solenoid valves for spray release through ﬂat-fan noz-

zles when mesquite canopies interrupted the light. The

levels of mesquite mortality achieved were equivalent

to those achieved by hand-spraying by ground crews.

The speed-sprayer (SS) has been widely used in

orchards and its autonomous control is a main theme

in the automation of the spraying operation. The elec-

tromagnetic induction type and pipe induction type

were commercialized in 1993 and 1994 [63.192,193];

both types require induction wires or pipes on the

ground, underground, or above ground at 150–200cm

height and between tree rows. Induction sensors, safety

sensors (ultrasonic sensor, touch sensors), and other in-

ternal sensors are installed in the unmanned SS,and

autonomous control is conducted withfuzzy theory. An-

other method of SS control with genetic algorithm and

fuzzy theory using GPS has been reported [63.194].

Figure 63.21 shows a multioperation robot with

a three-DOF Cartesian coordinate manipulator and an

end-effector [63.195]. When an end-effector shown in

Fig.63.22 is attached to the manipulator, it can weed

Part G 63.5

1118 Part G Infrastructure and Service Automation

Weeding

knife

40mm

DC motor

Fig. 63.22

A weeding end-

effector

on the ridge between crops. Color images of the weed

are fed to a computer from a color camera and the

three-dimensional location of the weed is calculated

using a binocular stereo method. Weed detection is con-

ducted using color or texture difference between weed

and soil or crops [63.196–198]. The end-effector was

a weed knife with a spiral shape (4cm diameter). This

robot can also be a leaf-vegetable harvesting robot or

a transplanting robot when the end-effectors were re-

placed [63.186].

63.5.2 Automation of Fruit Grading

and Sorting

Because of the ever-growing need to supply high-

quality food products within a short time, automated

grading of agricultural products is getting special prior-

Image B PCImage A PCX-ray image PC

X-ray camera Camera A out

Camera B out

Spin, 180° turn

X-ray generatorLight projector

Camera A in Camera A top DL light Camera B top

Judgement PC

Light interceptor

Sugar-acid PC

Camera B in

Fig. 63.23 A schematic diagram of the camera and lighting setup

ity among many farmers associations. The impetus for

these trends can be attributed to increased awareness of

consumers about their health well-being and a response

from producers to provide quality-guaranteed products

with consistency. It is in this context that the ﬁeld of au-

tomatic inspection and machine vision comes to play

the important role of quality control for agricultural

products [63.184–188]. Unlike most industrial prod-

ucts, quality inspection of agriculturalproducts presents

speciﬁc challenges because nonstandard products must

be inspected according to their appearance and inter-

nal quality, which are acceptable to customers only for

nondestructive methods [63.199]. Several sensors have

been developed and applied for internal quality deter-

mination, including sugar content, acidity, rind pufﬁng,

rotten core, and other internal defects [63.190–194].

Fruit Grading System with Conveyors

Figure 63.23 shows an automated inspection system for

quality control of various agricultural products, with

fruits and vegetables being the main ones. As a rep-

resentative of other agricultural products, discussion in

this section is focused mainly on the orange fruit, a ma-

jor agricultural product inspected by this system. The

main components of the system for automated inspec-

tion and sorting can be outlined as follows:

1. Product reception from supplier

2. Container unpacking and dumping of products

3. Feeding of products to the conveyor line

Part G 63.5

Automation in Agriculture 63.5 Fruit Production Operations 1119

4. Inspection for internal and external conditions and

defects followed by assignment of quality rating

5. Weight adjustment and release of the inspected

product into packing box

6. Labeling of grade and size using an inkjet printer

7. Box closure and sealing

8. Box transfer onto palette and loading ready for mar-

keting

These features areintegrated into an operational line

that combines advanced designs, expert fabrications,

and automatic mechanical control with the main ob-

jective of offering the best visual solutions and stable

quality judgment.

Illumination and Image-Capture Devices

Illumination is one of the most important components

for the machine-vision system to inspect products, be-

cause it determines the quality of images acquired,

especially for glossy products whose cuticular layers

are thick. Polarizing ﬁlters are sometimes used in front

of lighting devices and camera lenses to eliminate ha-

lation on the acquired images [63.195]. A color CCD

TV camera is often employed to sense light through

photosensitive semiconductor devices, and the CCD ar-

ray data is transferred by progressive scan mode to

the frame storage area, representing an image of the

scene [63.196]. The TV camera is equipped with one

chip that transfers red–green–blue (RGB) analogue data

and is set at a shutter speed of 1/1 000 s during inspec-

tion because line speed is usually 1m/s. Image-capture

boards with 8 bit level resolution and spatial resolution

of about 512×512 pixels have been used to store and

digitize video signals and output the data to computer

memory for analysis or display on the monitor. Re-

cently, a special image-acquisition device, a universal

serial bus (USB) or a local-area network (LAN) is often

used between TV camera and personal computer (PC)

instead of the image-capture board, enabling image pro-

cessing to be performed within 10–30ms.

Product Reception and Forwarding

The ﬁrst step in the inspection procedure starts at the

receiving platform, situated on the ground ﬂoor. Agri-

cultural products are packed into containers by the

farmers and delivered to the inspection factoryin trucks.

A folk lift is used to unload the containers on one pallet

and deliver them to the depalletizer device that sepa-

rates the containers automatically so that they are fed

one by one to the conveyor, which propels them to the

upper ﬂoor, where the main inspection line is located.

The depalletizer has a capacity to handle 1200–1400

pallets per hour. After depalletization, the containers are

handled by the dumper machine. The dumper is an au-

tomated machine that turns and empties the containers

gently and then spreads the fruit on a belt conveyor. Us-

ing specialized rollers fruits are singulated so that they

are fed singly to the roller-pin conveyor before process-

ing. To acquire a complete view of the fruit the roller

pins have been designed so that the fruit is always posi-

tioned at the center point.

Internal Quality Inspection

The ﬁrst stage in the inspection sequence is to deter-

mine the sugar and acid contents using a near-infrared

(NIR) inspection system. A special sensor determines

the sugar content (brix equivalent) and acidity level of

the fruits from light wavelengths received by speciﬁc

sensors after light is transmitted through the fruit. The

sensor photoelectrically converts the light into signals

and sends them to the computer unit, where they are

processed and classiﬁed. In addition, the internal fruit-

quality sensor measures the granulation level of the

fruit, which indicates its internal water content. Next,

the fruit is conveyed to the x-ray imaging component,

which inspects for biological defects such as rind puff-

ing and the granulation status of the juice sacs. X-ray

imaging operates by transmitting x-rays released from

a generator through the orange fruit. The emitted opti-

cal x-ray image is relayed to the x-ray scintillator, an

optical device that consists of a thin coat of lumines-

cent materials through which x-rays are converted into

the visible light of a normal image. The resulting im-

age is captured by a monochrome CCD camera and

copied to computer memory through an image-capture

board.

Image Analysis

At the next operation, the fruit is conveyed to the third

inspection stage where the main image processing and

grading takes place using factory automation comput-

ers (Fig.63.23). Six CCD cameras set in random trigger

mode acquire images of the fruits as they are conveyed

at constant speed. Firstly, two side cameras on the right

side, placed equidistance from the central position of

the fruit, capture images of the right side of the fruit.

Next, two side cameras on the left side, again placed

equidistance from the central position of the fruit, cap-

ture images of the left side of the fruit. Next, a top

camera acquires the top surface image. Finally, the fruit

is then spun through 180

◦

around a horizontal axis by

mechanically controlled rollerpins, which facilitatesthe

Part G 63.5

1120 Part G Infrastructure and Service Automation

acquisition of another image of the lower surface by

a sixth camera. Images are captured by CCD cameras

after a trigger signal is received through the digital in-

put/output (DIO) board. Sensors, wired to a sequencer

box, are used to track the fruit position and the output is

relayed to the DIO board from where the program reads

the 8bit output data.

Image processing is then executed and the following

features are inspected:

1. Size (maximum and minimum diameter, area, and

extrapolated diameter)

2. Color (color space based on hue, saturation, inten-

sity (HSI) values and RGB ratios)

3. Shape (perimeter L, L

/area, inﬂection point of

outer surface, and center of gravity)

4. Bruise (based on intensity of blue color level and

summation ofcolor levelsfor blue,and R–G derived

images)

Results of processing are written to a shared mem-

ory area using a memo-link from where a judgment PC

makes decision about the quality of the fruit. Grading

for quality is assigned by ranks of between four and

six. Graded fruitsare conveyed to the weight adjustment

machine, which controls the total weight of oranges to

be packed in a box according to preset values. Between

four and eight weight rankings of fruits are used to

ﬁll different pack boxes. An automatic barcode labeler

then prints the grade and size outputs on the box us-

ing an inkjet printer before the packing box is closed by

the packing machine and then automatically sealed by

the case-sealer machine. At the end of the product in-

spection line a robot-controlled palletizer completes the

process by arranging the boxes onto palettes ready for

onward loading onto trucks and transport to consumer

markets.

Data Maintenance

An important feature of the grading system design is

that it is adaptable to the inspection of different prod-

ucts such as potato, tomato, persimmon, sweet pepper,

waxed apple, and kiwi fruits with adjustments only

to the processing codes. Several lines for orange fruit

inspection combined with high conveyance and high-

speed microprocessors enable the system to handle

large batches of fruit product at high speeds. Apart from

quality inspection another objective of this system is to

gather data about product performance. Identiﬁcation of

certain defects and counts can lead to the discovery of

the cause and its severity. All data from receipt of fruits

at the collection site through the analytical processes,

packing, and shipping is stored in an ofﬁce computer

connected through a local-area network (LAN)tothe

corresponding factory automation PCs. Experienced

personnel manage the data, based on which day-to-day

operations can be monitored remotely. By making prod-

uct performance records available on the Internet it will

be possible to monitor performance online and provide

a useful service to thecustomers. Allin all, the customer

of a product is the ﬁnal judge of its quality. Therefore,

keeping internal standards and speciﬁcations in line

with customer expectations is a priority that is achieved

through good relationshipsand regular communications

with the customer.

Fruit Grading Robot

Based on the technologies of the grading system

described in the previous section and robotic tech-

nologies, a fruit grading robot as shown in Fig.63.24

was developed in 2002. This robot system has two

three-DOF Cartesian manipulators, 16 suction cups as

end-effectors, and 16 machine-vision systems consist-

ingof16colorTV cameras and 36 lighting devices.

This can be applied to tomato, peach, pear, apple, and

other fruits.

Operation ﬂow with this grading system is as fol-

lows:

1. When products are received, four blocks of contain-

ers (a block consists of ten containers) are loaded on

a pallet.

2. A block of the containers is lifted up to the second

ﬂoor, where the main parts of the grading system

work, and a container separator sends containers

one by one to a barcode reader.

Fig. 63.24 A tomato fruit grading robot

Part G 63.5

Automation in Agriculture 63.6 Summary 1121

3. After obtaining information from each barcode at-

tached to a container, the container is sent to a robot,

the fruit providing robot.

4. The robot sucks fruits up by using suction pads and

moves them to a halfway stage.

5. Another robot, the grading robot, picks fruits up

again from the halfway stage and bottom and side

images of fruits are acquired by TV cameras during

transferring fruits.

6. The robot transfers them to trays on a conveyor line

and a top image of the fruit is acquired in a camera

box.

7. After appearance inspection,internal conditions and

sugar content are inspected by an infrared analysis

sensor.

8. Fruit that pass the internal quality sensor box are

packed into a corrugated cardboard box by a pack-

ing robot based on their grading results. Grade, size,

and name of fruit variety are printed on the box sur-

face by an inkjet printer and the box is closed and

sealed.

9. Finally, the boxes are transferred onto palettes and

are loaded into a truck for marketing.

Figure 63.25 shows the actions of the two manipula-

tors: the fruit providing and grading robots. A container

in which 8×6, 6×5, 6×4 or 5×3 fruits are ﬁlled is

pushed into the working area of the providing robot

by a pusher (1). The providing robot has a three-DOF

Cartesian coordinate manipulatorand eightsuction pads

as end-effectors. The robot sucks eight (maximum)

fruits up (2) and transfers them to a halfway stage,

spacing fruitintervalsin the y direction (3). Two provid-

ing robots independently work and set 16 (maximum)

fruits on a halfway stage. A grading robot which con-

sists of another three-DOF manipulator (two prismatic

joints and a rotational joint) and 16 suction pads sucks

them up again (4) and moves them to trays on a con-

veyor line. Bottom images of fruits are acquired as the

grading robot moves over 16 color TV cameras. The

cameras and lighting devices turn down 90

◦

follow-

ing the grading robot’s motion (5). Before releasing the

Blower

Providing

robot

Pusher

Grading

robot

Halfway stage

TV camera

Lifter

Fig. 63.25 Actions of robots

fruits to trays on the line, the TV cameras acquire four

side images of fruits as they rotate through 270

◦

(6). Af-

ter image acquisition, the robot releases the fruits into

trays (7) and a pusher pushes 16 trays to a conveyor

line (8).

This gradingrobot’s maximum speedwas 1m/sand

its stroke was about 1.2m. It took 2.7s for the robot

to transfer the 16 fruit to trays, 0.4s to move down

from the initial position, 1s to move back from re-

leasing fruits, and 0.15s for waiting. Total time was

4.25s to move back and forth for the stroke. This makes

this robot performance approximately 10000fruit/h. In

this system, four blowers with speciﬁcation of 1.4kW,

3400rpm, 38kPa, 1.3m

/min displacement were used

for two providing robots and a grading robot. About

30kPa vacuum force was suitable for sucking peach

fruit, while 45kPa was used for pear and apple fruits

and no damage was observed even after sucking peach

fruits twice. The tray has a data carrier (256byte

EE-PROM), and grading information of each fruit is

sent from a computer to the data carrier through an an-

tenna after image processing. A conveyor line transfers

the trays at 30m/min [63.197].

63.6 Summary

Despite the problems in introducing automation into

agricultural production systems many automation sys-

tems have been developed and are commonly applied in

agricultural operations. Automation has increased the

efﬁciency and quality of agricultural production sys-

tems.

However, automated or semiautomated farming is

far from a reality in many parts of the world. Due to

Part G 63.6

1122 Part G Infrastructure and Service Automation

cheap labor in third-world countries, much of the work

on farms is still performed manually. Despite the large

capital investment needed to purchase the equipment,

automation will probably be introduced also into these

countries to provide the needs for increased production

and land efﬁciency.

In industrialized countries the production trend is

towards large-scale farms and hence automation will

be advanced and commercialized to make this feasible.

Farmers must produce their food at competitive prices

to stay in business and automation of farming technol-

ogy is the only way forward. With the improvement

of sensors and computers, and decrease of automation

equipment costs, this is becoming feasible and more

systems will be introduced. The current century will

probably see signiﬁcant advances in automation and

robotization of farm operations. The future farm will

include integration of advanced sensors, controls, and

intelligent software to provide viable solutions to the

complex agricultural environment.

References

63.1 J.K. Schueller: Automation and control. In: CIGR

Handbook of Agricultural Engineering,Informa-

tion Technology, Vol. VI, ed. by A. Munack (CIGR,

Tzukuba 2006) pp. 184–195, Chap. 4

63.2 H.G. Ferguson: Apparatus for coupling agricultural

implements to tractors and automatically regulat-

ing the depth of work, Patent GB 253566 (1925)

63.3 G. Singh: Farm machinery. In: Agricultural Mech-

anization & Automation, Encyclopedia of Life

Support Systems (EOLSS),ed.byP.McNulty,

P.M. Grace (EOLSS, Oxford 2002)

63.4 J.N. Wilson: Guidance of agricultural vehicles -

a historical perspective, Comput. Electron. Agric.

25(1), 3–9 (2000)

63.5 T. Torii: Research in autonomous agriculture vehi-

cles in Japan, Comput. Electron. Agric. 25(1), 133–153

(2000)

63.6 R. Keicher, H. Seufert: Automatic guidance for

agricultural vehicles in Europe, Comput. Electron.

Agric. 25(1), 169–194 (2000)

63.7 J.F.Reid,Q.Zhang,N.Noguchi,M.Dickson:Agri-

cultural automatic guidance research in North

America, Comput. Electron. Agric. 25(1), 155–167

(2000)

63.8 H. Auernhammer, T. Muhr: GPS in a basic rule for

environment protection in agriculture, Proc. Au-

tom. Agric. 11(91), 394–402 (1991)

63.9 M. O’Connor, T. Bell, G. Elkaim, B. Parkinson: Auto-

matic steering of farm vehicles using GPS, Proc. 3rd

Int. Conf. Precis. Agric. (Minneapolis 1996) pp. 767–

778

63.10 T. Stombaugh, E. Benson, J.W. Hummel: Auto-

matic guidance of agricultural vehicles at high ﬁeld

speeds, ASAE Paper No. 983110 (ASAE, St. Joseph

1998)

63.11 T. Bell: Automatic tractor guidance using carrier-

phase differential GPS, Comput. Electron. Agric.

25(1/2), 53–66 (2000)

63.12 N. Noguchi, M. Kise, K. Ishii, H. Terao: Field

automation using robot tractor, Automation Tech-

nology for Off-road Equipment, Proc. 26–27 July

Conf., ed. by Q. Zhang (ASAE, Chicago 2002) pp. 239–

245

63.13 G.P. Gordon, R.G. Holmes: Laser positioning sys-

tem for off-road vehicles, ASAE Paper No. 88-1603

(ASAE, St. Joseph 1988)

63.14 N. Noguchi, K. Ishii, H. Terrao: Development of an

agricultural mobile robot using a geomagnetic di-

rection sensor and image sensors, J. Agric. Eng.

Res. 67, 1–15 (1997)

63.15 J.F. Reid, S.W. Searcy, R.J. Babowic: Determining

a guidance directrix in row crop images, ASAE Paper

No. 85-3549 (ASAE, St. Joseph 1985)

63.16 J.B. Gerrish, G.C. Stockman, L. Mann, G. Hu: Image

rocessing for path-ﬁnding in agricultural ﬁeld op-

erations, ASAE Paper No. 853037 (ASAE, St. Joseph

1985)

63.17 J.A. Marchant, R. Brivot: Real time tracking of plant

rows using a Hough transform, Real Time Imaging

1, 363–375 (1995)

63.18 J.A. Marchant: Tracking of row structure in three

crops using image analysis, Comput. Electron.

Agric. 15, 161–179 (1996)

63.19 S. Han, Q. Zhang, B. Ni, J.F. Reid: A guidance

directrix approach to vision-based vehicle guid-

ance systems, Comput. Electron. Agric. 43,179–195

(2004)

63.20 J. Billingsley, M. Schoenﬁsch: The successful

development of a vision guidance system for agri-

culture, Comput. Electron. Agric. 16(2), 147–163

(1997)

63.21 J.A. Farrell, T.D. Givargis, M.J. Barth: Real-time dif-

ferential carrier phase GPS-aided INS, IEEE Trans.

Control Sys. Technol. 8(4), 709–721 (2000)

63.22 L. Guo, Q. Zhang, S. Han: Position estimate of off-

road vehicles using a low-cost GPS and IMU, ASAE

Paper No. 021157 (ASAE, St. Joseph 2002)

63.23 M.A. Abidi, R.C. Gonzales: Data fusion. In: Robotics

and Machine Intelligence (Academic, San Diego

1992)

63.24 M.S. Grewal, A.P. Andrews: Kalman Filter: Theory

and Practice Using MATLAB, 2nd edn. (Wiley, New

York 2001)

63.25 O. Cohen, Y. Edan: A new framework for online

sensor and algorithm selection, Robot. Auton. Syst.

56(9), 762–776 (2008)

Part G 63

Automation in Agriculture References 1123

63.26 H. Choset: Coverage for robotics - a survey of recent

results, Ann. Math. Artif. Intell. 31, 113–126 (2001)

63.27 T. Oksanen, S. Kosonen, A. Visala: Path planning

algorithm for ﬁeld trafﬁc, ASAE Paper No. 053087

(ASAE, St. Joseph 2005)

63.28 J. Jin, L. Tang: Optimal path planning for arable

farming, ASAE Paper No. 061158 (ASAE, St. Joseph

2006)

63.29 U. Shani: Filling regions in binary raster images:

a graph-theoretic approach, SIGGRAPH ’80 Conf.

Proc. (ACM, New York 1980) pp. 321–327

63.30 Y.Y. Huang, Z.L. Cao, E.L. Hall: Region ﬁlling oper-

ations for mobile robot using computer graphics,

Proc. of IEEE Int. Conf. Robot. Autom. (1986)

pp. 1607–1614

63.31 Z.L. Cao, Y. Huang, E.L. Hall: Region ﬁlling opera-

tions with random obstacle avoidance for mobile

robots, J. Robot. Syst. 5(2), 87–102 (1988)

63.32 S.A. Gray: Planning and Replanning Events for Au-

tonomous Orchard Tractors. Ph.D. Thesis (Utah State

University, Utah 2001)

63.33 J. Park, P.E. Nikravesh: A look-ahead driver model

for autonomous cruising on highways, 1996 Future

Transport. Technol. Conf. Expo. (Warrendale, 1996)

63.34 Q. Zhang, H. Qiu: A dynamic path search algorithm

for tractor automatic navigation, Trans. ASAE. 47(2),

639–646 (2004)

63.35 D. Wu, Q. Zhang, J.F. Reid, H. Qiu: Adaptive control

of electrohydraulic steering system for wheel-type

agricultural tractors, ASAE Paper No. 993079 (ASAE,

St. Joseph 1999)

63.36 Q. Zhang: Hydraulic linear actuator velocity control

using a feedforward-plus-PID control, Int. J. Flex.

Autom. Integr. Manuf. 77, 275–290 (1999)

63.37 H. Qiu, Q. Zhang, J.F. Reid: Fuzzy control of electro-

hydraulic steering systems for agricultural vehicles,

Trans. ASAE 44(6), 1397–1402 (2001)

63.38 L. Guo, Q. Zhang, S. Han: Agricultural machin-

ery safety alert system using ultrasonic sensors, J.

Agric. Saf. Health 8(4), 385–396 (2002)

63.39 J. Wei, F. Rovira-Mas, J.F. Reid, S. Han: Obstacle

detection using stereo vision to enhance safety of

autonomous machines, Trans. ASAE 48(6), 2389–

2397 (2005)

63.40 M. Kise, Q. Zhang, N. Noguchi: An obstacle identi-

ﬁcation algorithm for a laser range ﬁnder-based

obstacle detector, Trans. ASAE 48(3), 1269–1278

(2005)

63.41 Y. Matsuo, S. Yamamoto, O. Yukumoto: Develop-

ment of tilling robot and operation software. In:

Autom. Technol. Off-Road Equip. (ATOE) Proc., ed.

by Q.Zhang, ASAE Publication No. 701P0509 (2002)

pp. 184–189

63.42 J.F. Reid: Mobile intelligent equipment for off-road

environments, Proc. ATOE Conf. (ASAE, St. Joseph

2004) pp. 1–9

63.43 T. Pilarski, M. Happold, H. Pangels, M. Ollis,

K. Fitzpatrick, A. Stentz: The Demeter system for

automated harvesting, Proc. 8th Int. Top. Meet.

Robot. Remote Syst. (1999)

63.44 T. Pilarski, M. Happold, H. Pangels, M. Ollis,

K. Fitzpatrick, A. Stentz: The Demeter system for

automated harvesting, Auton. Robot 13,19–20

(2002)

63.45 RAS: Robotics and Automation Society,Service

Robots (IEEE, Piscataway 2008)

63.46 B. Astrand, A.J. Baerveldt: An agricultural mobile

robot with vision-based perception for mechanical

weed control, Auton. Robot 13, 21–35 (2002)

63.47 R.N. Jørgensen, C.G. Sørensen, J.M. Pedersen,

I. Havn, H.J. Olsen, H.T. Søgaard: Hortibot: An

accessory kit transforming a slope mower into

a robotic tool carrier for high-tech plant nursing

- part I, ASAE Paper No. 63082 (ASAE, St. Joseph

2006)

63.48 C.G. Sørensen, R.N. Jørgensen, M. Nørremark: Hort-

iBot: Application of quality function deployment

(QFD) method for horticultural robotic tool carrier

design planning - part II, ASAE Paper No. 67021

(ASAE, St. Joseph 2006)

63.49 M. Nørremark, C.G. Sørensen, R.N. Jørgensen:

HortiBot: Comparison of potential present and fu-

ture weeding technologies - part III, ASAE Paper

No. 67023 (ASAE, St. Joseph 2006)

63.50 J.L. Merriam, S.W. Styles, B.J. Freeman: Flexible ir-

rigation systems: concept, design, and application,

J. Irrig. Drain. Engrg. 133(1), 2–11 (2007)

63.51 S.J. Kim, P.S. Kim: Optimal gate operation of irriga-

tion reservoir using water management program,

ASAE Paper No. 042067 (ASAE, St. Joseph 2004)

63.52 G. Park, M.S. Lee, S.J. Kim: Networking model of

paddy irrigation system using archyhydro GIS, ASAE

Paper No. 052079 (ASAE, St. Joseph 2005)

63.53 Y. Lam, D.C. Slaughter, W.W. Wallender, S.K. Upad-

hyaya: Machine vision monitoring for control of

water advance in furrow irrigation, Trans. ASAE

50(2), 371–378 (2007)

63.54 Y. Lam, D.C. Slaughter, S.K. Upadhyaya: Computer

vision system for automatic control of precision

furrow irrigation system, ASAE Paper No. 062078

(ASAE, St. Joseph 2006)

63.55 Y. Kim, R.G. Evans, W. Iversen, F.P. Pierce,

J.L. Chavez: Software design for wireless in-ﬁeld

sensor based irrigation management, ASAE Paper

No. 063704 (ASAE, St. Joseph 2006)

63.56 N.L. Klocke, C. Hunter, M. Alam: Application of

a linear move sprinkler system for limited irrigation

research, ASAE Paper No. 032012 (ASAE, St. Joseph

2003)

63.57 B.A. King, R.W. Wall, L.R. Wall: Distributed

control and data acquisition system for closed-

loop site-speciﬁc irrigation management with

center pivots, Appl. Eng. Agric. 21(5), 871–878

(2005)

63.58 M. Yitayew, K. Didan, C. Reynolds: Microcomputer

based low-head gravity-ﬂow bubbler irrigation

Part G 63

1124 Part G Infrastructure and Service Automation

system design, Comput. Electron. Agric. 22,29–39

(1999)

63.59 F.S. Zazueta, A.G. Smajstrla: Microcomputer-based

control of irrigation systems, Appl. Eng. in Agric.

8(5), 593–596 (1992)

63.60 B. Cardenas-Lailhacar, M.D. Dukes, G.L. Miller:

Sensor-based control of irrigation in Bermuda-

grass, ASAE Paper No. 052180 (ASAE, St. Joseph

2005)

63.61 M.B. Haley, M.D. Dukes: Evaluation of sensor based

residential irrigation water application, ASAE Paper

No. 072251 (ASAE, St. Joseph 2007)

63.62 S.R. Evett, R.T. Peters, T.A. Howell: Controlling

water use efﬁciency with irrigation automation,

South. Conserv. Syst. Conf. (Amarillo 2006)

63.63 D.F. Wanjuru, S.J. Maas, J.C. Winslow, D.R. Up-

church: Scanned and spot measured temperatures

of cotton and corn, Comput. Electron. Agric. 44,

33–48 (2004)

63.64 S.R. Herwitz, L.F. Johnson, S.E. Dunagan, R.G. Hig-

gins, D.V. Sullivan, J. Zheng, B.M. Lobitz, J.G. Le-

ung, B.A. Gallmeyer, M. Aoyagi, R.E. Slye, J.A. Brass:

Imaging from an unmanned aerial vehicle surveil-

lance and decision support, Comput. Electron.

Agric. 44, 49–61 (2004)

63.65 J.A. Poss, W.B. Russell, P.J. Shouse, R.S. Austin,

S.R. Grattan, C.M. Grieve, J.J. Lieth, L. Zheng:

A volumetric lysemeter system: an alternative

to weighing lysimeters for plant-water rela-

tions studies, Comput. Electron. Agric. 43, 55–68

(2004)

63.66 Y. Kim, R.G. Evans, W. Iversen, F.P. Pierce: Instru-

mentation and control for wireless sensor network

for automated irrigation, ASAE Paper No. 061105

(ASAE, St. Joseph 2006)

63.67 T. Hess: A microcomputer scheduling program for

supplementary irrigation, Comput. Electron. Agric.

15, 233–243 (1996)

63.68 M.J. Upcraft, D.H. Noble, M.K.V. Carr: A mixed linear

programme for short-term irrigation scheduling, J.

Oper. Res. Soc. 40(10), 923–931 (1989)

63.69 K. Milla, S. Kish: A low cost microprocessor and

infrared sensor system for automating water in-

ﬁltration measurements, Comput. Electron. Agric.

53, 122–129 (2006)

63.70 J. Artigas, A. Beltran, C. Jimenez, A. Baldi, R. Mas,

C. Dominguez, J. Alonmso: Application of ion sen-

sitive ﬁeld effect transistor based sensor for soil

analysis, Comput. Electron. Agric. 31(3), 281–293

(2001)

63.71 R.T. Peters, S.R. Evett: Using low-cost GPS receivers

for determining ﬁeld position of mechanized ir-

rigation systems, Appl. Eng. Agric. 21(5), 841–845

(2005)

63.72 Y. Kim, R.G. Evans, W. Iversen, F.P. Pierce: Evalua-

tion of wireless control for variable rate irrigation,

ASAE Paper No. 062164 (ASAE, St. Joseph 2006)

63.73 F.R. Miranda, R. Yoder, J.B. Wilkerson: A site-

speciﬁc irrigation control system, ASAE Paper No.

031129 (ASAE, St. Joseph 2003)

63.74 F.R. Miranda, R.E. Yoder, J.B. Wilkerson, L.O. Odhi-

ambo: An autonomous controller for site-speciﬁc

management of ﬁxed irrigation systems, Comput.

Electron. Agric. 468, 183–197 (2005)

63.75 King B.A. W.W. Wall, D.C. Kincaid, D.T. Wester-

mann: Field testing of a variable rate sprinkler and

control system for site-speciﬁc water and nutrient

application. Appl. Eng. Agric.

21(5), 847–853 (2005)

63.76 A.T. Csordas, M.J. Delwiche, J. Barak: Automated

real-time PCR Biosensor for the detection of

pathogens in produce irrigation water, ASAE Paper

No. 047045 (ASAE, St. Joseph 2004)

63.77 N. Kondo, K.C. Ting (Eds.): Robotics for Bioproduc-

tion Systems (ASAE, St. Joseph 1998)

63.78 T. Mitsuhashi, A. Yamazaki, T. Shichishima: Au-

tomation of plant factory, Proc. 4th SHITA Symp.

(Tokyo 1994) pp. 45–57

63.79 N.Kondo,M.Monta,N.Noguchi:Agri-Robots (II)

Mechanisms and Practice (Corona, Tokyo 2006)

pp. 1–223

63.80 W. Simonton: Automatic geranium stock processing

in a robotic workcell, Trans. ASAE 33(6), 2074–2080

(1990)

63.81 N. Kondo, M. Monta: Basic study on chrysanthe-

mum cutting sticking robot, Proc. Int. Symp. Agric.

Mech. Autom., Vol. 1 (1997) pp. 93–98

63.82 N. Kondo, M. Monta, Y. Ogawa: Cutting pro-

viding system and vision algorithm for robotic

chrysanthemum cutting sticking system, Preprints

of the International Workshop on Robotics and Au-

tomated Machinery for Bioproductions (Valencia

1997) pp. 7–12

63.83 U-shin LTD.: US-500 Users manual (Tokyo, 1993)

63.84 E. Nederhoff: Energy and CO

Enrichment (Galileo

Services Ltd, New Zealand 2007), http://www.

redpathaghort.com/bulletins/co2.html

63.85 C. Kittasa, N. Katsoulasa, A. Bailleb: SE-structures

and environment: Inﬂuence of greenhouse ven-

tilation regime on the microclimate and energy

partitioning of a rose canopy during summer con-

ditions, J. Agric. Eng. Res. 79(3), 349–360 (2001)

63.86 E.J. van Henten: Greenhouse climate manage-

ment: an optimal control approach. Ph.D. Thesis

(Wageningen University, Holland 1994)

63.87 R. Caponetto, L. Fortuna, G. Nunnari, L. Occhipinti,

M.G. Xibilia: Soft computing for greenhouse cli-

mate control, IEEE Trans. Fuzzy Sys. 8(6), 1101–1120

(2000)

63.88 T. Morimoto, Y. Hashimoto: An intelligent control

for greenhouse automation, orieneted by the con-

cepts of SPA and SFA, Comput. Electron. Agric. 29,

3–20 (2000)

63.89 L.D. Albright: Controlling greenhouse environ-

ments, Acta Horticulturae 578, 47–54 (2002)

Part G 63