Technical Summary:
This proposal is a
research to explore the feasibility of merging in-pond assets with the
ubiquitous information technology of the farm office. It will also explore
the feasibility of lowering the cost of manufacturing and using GPS-based
equipment so that more site specific automation can be done economically.
The project will apply industrial standard wireless information network
technology to create an extensible and robust infrastructure that will
bridge different information sources and control points on the farm.
The proposed project,
if implemented, will create an information infrastructure for instruments in
the cabs of equipments, other sensors, and the farmer’s information system.
It will enhance the movement of data between in-farm assets and other
information systems to a point that they look like web sites to the users
with the data exchange between them automated and transparent. I addition,
the availability of error correction for Global Positioning System (GPS)
receivers will become free from the recurring subscription cost. This will
facilitate the development and the use of lower cost non-differential
GPS-based systems without sacrificing much accuracy to make precision
agriculture more feasible.
The proposed project,
if successful, will impact the practice of site-specific agriculture in at
least two ways: 1. Reduce the cost of manufacturing and using of many
GPS-based equipment thus enabling more site-specific automation, 2. Reduce
the cost and simplify the management of electronic information on the farm.
Objectives:
The objectives of this project are:
1.
Develop and implement a prototype wireless information
infrastructure.
2.
Investigate the feasibility and efficacy of the implementation in an
environment of a robotic in-pond aquacultural monitoring system using sonar
based sensors as a model to monitor pond inventory.
Procedures:
1.
Develop and implement a prototype wireless information
infrastructure. Figure 1 shows the basic architecture of the infrastructure
in a simplified form. The Base Station block is an embedded computer system
packaged for extended outdoor operation. The configuration of the Base
Station will include a GPS receiver, a wireless transceiver which serves as
an access point (AP) to which other wireless node can communicate. It
contains a minimal amount of data storage capability, and it contains the
software necessary to serve the other blocks in the infrastructure as well
as serving as a relay to another base station block.
|
Figure 1. Overall
architecture of the proposed information infrastructure |
The Sensor and
Equipment block is an embedded computer packaged to operate in a floating
platform. The platform is designed to accommodate various sensor and control
packages to make it readily interface to other ASTA projects. The embedded
computer is configured to have a GPS receiver, a wireless transceiver that
works as a station adaptor (SA) for communication to an AP. The platform is
self-contained in that it holds a battery power source as well as a
pump-based propulsion system for it to navigate autonomously.
The User block
represents other wireless data source or data consumer. It can be another
Base Station block or another set of the basic infrastructure.
The construction of
the above mentioned basic infrastructure was completed in the 2001-2002
project year, where one floating platform was constructed, a laptop computer
was used as the user block and a basic web-based user interface was
implemented (see progress report).
The tasks planned to
complete this objective is to focus on the modifications and refinement of
the hardware and software to facilitate better assessment of its efficacy.
An addition floating platform will be constructed using a different
flotation technology to facilitate the study of different data traffic
scenarios. A portable Base Station Block will be constructed to facilitate
the study of scalability.
2.
Investigate the feasibility and efficacy of the implementation in the
environment of a robotic in-pond monitoring system.
A sonar based sensor system for monitoring pond inventory will be used as a
model system to generate information traffic for testing the robustness of
the infrastructure. The data traffic environment will include long messages,
multiple simultaneous transmissions, and loss connection conditions.
Enhancement and modifications of the system and application software will be
done to include viewing of multiple ponds, and the access of pond
information in the South Farm of MSU campus from the Biological Engineering
Department in a multi-hop arrangement (using the portable Base Station).
The kind of result
expected from this project:
The expected outcomes of the proposed
project include:
The following milestones will be used to
indicate the success of the project:
- The development
of operation of a web based control system
- The development
of adapters that will add web-based capability to instruments
- The collection
of aquaculture data (Robinson and Wise) using the facility developed by
this project.
Techniques to be
employed, including their feasibility
This phase of the project will construct an additional floating platform and
a portable Base Station using the experiences learned from the previous year
to facilitate a more comprehensive data transfer scenarios.
Pitfalls which might
be encountered
The learning curve for graduate assistant to become familiar with this
system is significant. We lost the one graduate student hired for this
project to other academic department last year in the mid point of the
project. We will hire a new graduate student and get him/her onboard as
early as possible. Our contingency plan is to hire an additional
undergraduate student and let him/her work in parallel with the graduate
assistant for redundancy purposes. We realize that there is not fool-proof
contingency plan, and we will have to live with reality.
Limitations to
proposed procedure
Although the proposed project has a wide range of applications, the limited
scope of the project will provide ample opportunities for investigating the
concepts. The proposed project will initially be useable for research in
aquaculture only.
Justification:
American agriculture
is at the forefront of global competitiveness mainly due to its unparalleled
efficiency brought about by the high degree of labor saving
automation/mechanization. Data transport and management are an essential
part of modern agriculture. Automation is an obvious choice for improving
efficiency and reducing costs for the long run, and automation is inherently
linked tightly to data. The amount of automation implemented on farms and in
farm equipment has increased over the years and will likely increase
exponentially in the future. It has also become more and more a reality that
today’s commercial farms are research sites as well, which is synonymous to
the increase need for efficient data management.
Many farmers collect
and use data from almost all aspects of their farm operations to allow them
to maximize profit margin. The sources of data include but not limited to:
mechanized equipment, instruments, scouts, and the market. The amount of
data that must be communicated and managed, if not handled efficiently, can
become a costly aspect of a farm operation. Equipment and information
automation is becoming crucial for the success of modern agriculture, and
for the maintenance of our competitiveness in the global market. Damaged,
bad, or untimely data can be detrimental to the success of a modern
agricultural operation on the other hand good and reliable timely data can
ultimately help the bottom line.
Free differential correction signal source provided by the coastguard cannot
be reliably received under many circumstances it is highly susceptible to
interference and its coverage is not statewide. Subscribed satellite
correction system on the other hand has recurring cost that can be
substantial if the number of receivers on the farm is large. Despite the
removal of selective availability (SA) on GPS signal in 2000, differential
GPS is still needed in may agricultural operations that require a higher
degree of precision. The cost difference between a differential GPS receiver
and a non-differential GPS receiver is substantial. For example the cost of
a differential GPS receiver used in another ASTA research project ranges
from $3,000 to $3,500, while the cost of a non-differential GPS receiver
ranges from $200 to $400. Furthermore, the cost of OEM non-differential GPS
receiver ranges from $50 to $100. The substantial cost difference in the GPS
receiver alone begs for an alternative method of obtaining accurate
geo-spatial data. A base station technique of differential correction is low
cost because it is software-based. While it is not the fastest way of doing
correction (as opposed to hardware), it is adequate for most site-specific
agricultural applications. The impact of lower cost GPS-based equipment for
site-specific agriculture can be very significant. It will not only make a
wide spread use of site-specific automated equipment feasible, it has the
potential of improving the bottom line of production agriculture as a whole,
because of the improved efficiency the can be achieved with precision
agricultural practice.
Distributed data
sources are the inherent nature of an agricultural environment. The
relatively large spatial separation between different sources of data and
the number of such data sources that may exist in a given farming operation
present a challenge for the users who collect data from them or monitor
them. The challenges include:
-
Frequent
plug and unplugging of data cards which causes mechanical wear that can
make delicate equipment to fail.
-
The need to
retrieve, and store data in a frequent basis.
-
The involvement of
unskilled (inadequately trained) workers in the use of automated
equipment.
-
Shielding the communication and
management of data from the user is important because it eliminates
problems associated with human errors. The high degree of data automation
that will be achievable using the base station approach proposed in this
project will fulfill this need.
Why use ponds as demonstration site?
What are the other applications of this
project?
The proposed project will create a basis by which a scalable information
infrastructure can be built. Any on-farm instrument with a communication
capability will be adaptable to use the system proposed here. Example farm
applications that can benefits from this project include: machinery
management, environmental monitoring, scouting, harvesting, etc. We cannot
include many applications as a modeling system for this project at this
time, because we have to focus on creating the system.
Is this a reinventing of wheels?
This project uses many off-the-shelf components that are available in the
commercial market in different forms, but they are not suitable for on-farm
environment because some components are not made for outdoor or unattended
environment. What makes it unique is that it focuses on a low cost approach
that integrates a minimal data storage and data transport, and it can
support different communication protocols, which is a necessity in
agricultural environments. A configuration with multiple units will result
in a robust redundant infrastructure that is very desirable in agriculture.
The proposed project will create a "nugget" that can be replicated to form
the information infrastructure that is suitable for on-farm applications. It
contains an information server that provides differential correction
information to other GPS-based equipment on demand to make it possible to
use low-cost non-differential GPS systems in instruments and give an
acceptable accuracy without the need of subscription fees. This will impact
the cost of GPS based instruments in general. Is it the hardware, software,
or the knowledge that is important? All of the above. We need to develop the
knowledge to make this "easy" and "cheap", we need to develop software, and
we need to create turn-key hardware.
Literature Review:
Commercial wireless communication systems that will be used for this
research are those that are based on spread spectrum, frequency hopping
technologies. This mode of wireless communication is suitable for
agricultural use because of their relative immunity to radio interference.
Current research in the area of wireless data communication include the
activities at the Mobile Robot Laboratory of Georgia Tech, and Carnegie
Mellon University’s Robotic Engineering Laboratory, where autonomous robots
are being developed for various applications. Commercial products from Aries
Robot, WI, does driverless tractor for sewage maintenance. The use of a
wireless modem for controlling a waste water dispersing gun is an ongoing
ASTA project at Mississippi State University. The use of a base station for
differential correction is an established technique, commercial base station
systems are available, but they are sold in a packaged form that defeats the
purpose of low-cost implementation, furthermore they all use proprietary
software that limits the functionality of a base station to GPS only. Recent
report presented by McKinnion (USDA ARS, Beltwide Cotton Conference 2002) on
the use of wireless network on the farm had some similar idea as this
project, but the focus was geared toward the use of DIRECT PC technology for
cotton research.
Current Research:
Research conducted last year has produced a working prototype consisting of
a Base Station, a Sensor Platform (floating platform), and some initial
web-based software. The experiences gained from the work have helped us to
consider several modifications and improvements. Among the shortcomings
discovered from the previous year are: The floating platform must be
designed with as minimal number of features as possible to allow for better
power management and better choices of materials. Multiple floating
platforms are necessary to facilitate more extensive communication
scenarios.
Back to Top
Back to
Menu
|
