Agriculture Reference
In-Depth Information
In addition, the sensors must have an adequate range to follow the measured
parameter from its lowest value to its highest value. The sensor must also have ade-
quate resolution to detect significant differences in the sensed quantity. And the dif-
ference between precision and accuracy must be kept in mind.
In many cases, the sensors used for other applications can be used for agricultural
automation. Although there are many exceptions, the requirements for agricultural
automation with respect to accuracy, range, and resolution usually do not exceed
those of other sectors of the economy. However, the environment, whether outdoors
or in what agriculture considers a controlled environment, is often severe. The sen-
sors need to be reliable under such conditions.
Of even more concern is the variety of factors and complex heterogeneity of agri-
cultural objects. Agricultural objects are often complex combinations of physical,
chemical, and biological characteristics. For example, measuring the moisture con-
tent of a plant component or the fat content of a part of a live animal sounds simple,
but they are very complex tasks with many other varying parameters that can cause
sensors to give false readings. In many of the following chapters, there are extended
discussions of sensors and sensor development. Lessons from these experiences
should be used when designing an agricultural automation system.
Automation system variables may also need to be sensed in many systems. That
is, the output or some intermediate quantity may need to be measured to improve
the performance of the system. This is often necessary to deal with parameter varia-
tions within the agricultural automation system or to counteract disturbances on the
system. Typically, physical quantities, such as flow rates, displacements, and tem-
peratures are measured using sensors common to a wide variety of agricultural and
nonagricultural industries.
The many different types of agricultural systems mean that there are many dif-
ferent quantities to be sensed. And each of these quantities often has a variety of
sensing methods and sensor types. Some examples of quantities to be sensed and
potential sensors include:
r Displacement: potentiometers, LVDTs, capacitive sensors, encoders
r Velocity: DC tachometers, variable reluctance sensors, Hall effect sensors
r Temperature: thermocouples, thermistors, RTDs
r Moisture: conductance, capacitance, near-infrared spectroscopy
r Pressure: strain gauge diaphragm, piezoelectric
r Flow: venturi, turbine, hot wire anemometer, vortex shedding, coriolis
Again, the later chapters give additional examples.
A recent trend in agricultural sensors has been the increasing use of noncontact
spectral and vision sensors. Spectral sensors measure the emission, transmission,
reflectance, or absorbance of particular frequencies of electromagnetic radiation.
They are particularly effective at determining constituents and quality. Sensors out-
side the visible band, such as those using near-infrared, far-infrared, ultraviolet,
microwave, or terahertz bands, have become common.
Advances in machine vision, including better computational capabilities as well
as improved vision sensors with more resolution and sensitivity, have led to the wider
