Continuously Operating Reference Station (CORS) GNSS networks A by knm75792

VIEWS: 150 PAGES: 10

									   Continuously Operating Reference Station (CORS) GNSS networks: A
            Superior Infrastructure for Precision Agriculture

                                     Craig Roberts
                 School of Surveying and Spatial Information Systems,
                  University of New South Wales, Sydney, Australia
                                c.roberts@unsw.edu.au


Introduction

Global Navigation Satellite Systems (GNSS) and the most prominent member of this
family, the Global Positioning System (GPS), offer great economic benefits for the
agricultural sector. GPS was initially developed by the US Department of Defence
primarily to provide military advantage over adversaries, however in recent years,
civilian users out number military users by a thousand to one as GPS in mobile phones
and car navigation systems become mainstream. Nowadays the Russian system, Glonass
and future systems from Europe (Galileo) and China (Compass) complete the GNSS
family.

Early precision applications of GPS were for scientific purposes to measure and monitor
plate tectonic movements. In the early 1990s national governments, Australia included,
began to redefine their national coordinate frameworks using permanently mounted GPS
stations. This new framework in Australia is called the Geocentric Datum of Australia
(GDA94) and is compatible with modern satellite based positioning techniques such as
GNSS. Similarly state government land authorities have also been establishing their own
continuously operating reference station networks (CORS) which are aligned with the
GDA94 national coordinate framework.

A distinction between the quality of CORS networks was categorized in Rizos (2008) as
Tier 1, 2 and 3 stations. Tier 1 stations are the IGS stations (designed for scientific and
geodetic purposes), Tier 2 the primary national geodetic network (designed for datum
maintenance) and Tier 3 state managed and/or private networks (used for a range of
applications). These state based CORS networks were initially set up to reduce the cost of
survey mark maintenance, but are now considered a positioning infrastructure for
applications including a range of surveying operations as well as machine guidance in
mining and construction, precision agriculture, emergency management and asset
management. This paper will focus on the benefits of the growing CORS infrastructure
for precision agriculture.

Current status of CORS in Australia

The evolution of state based CORS networks in Australia has been ad hoc. Victoria was
the first state to establish a CORS network in 1994. They adopted a cooperative approach
and a mix of contributing stakeholders and equipment and now boast the only statewide
CORS network in Australia called VICmap Position GPSnet. GPSnet provides sub metre
accuracy, real-time positioning across the entire state with a denser sub network clustered
around Melbourne and environs delivering 2 cm accuracy, network RTK positions in
real-time.




Figure 1 – Vicmap Position GPSnet. Note the ochre shaded areas which deliver real-time
                  network RTK positions to users at 2cm precision.

In 2001 Queensland established a Trimble only VRS network around Brisbane and
environs. It has since grown to eight stations servicing SE Queensland. NSW took a more
corporate approach and established SydNet in 2004, testing carefully before releasing it
for use by the positioning community. This has since evolved to a 23 station network and
will ultimately cover the state as CORSnet NSW offering a mix of equipment and
network RTK capabilities.

GPSNet Perth, established in 2006, was the first privately owned CORS network in
Australia and now comprises eleven stations covering the city of Perth and environs.

Other smaller networks are also evolving such as the four station Darwin network, two
stations in Alice Springs, two state funded stations in Tasmania and one station in
Adelaide which is an expansion of the Victorian GPSnet in the neighbouring state.
Figure 2 – CORSnet NSW showing current and future stations across NSW. Currently
                      23 operating stations (July 2009)




           Figure 3 – GPSnet Perth (left) and Figure 4 - SunPoz (right)
Much of this recent activity has been encouraged by the Federal government granting
$AUS100M under the Auscope project to rollout 100 new GNSS CORS stations to
improve the accuracy of the national geodetic network. State governments have provided
matching funding to maximize the coverage over such vast areas of land.




                         Figure 5 – Proposed AuScope network

GPS manufacturers and their Australian distributors are also establishing their own
private CORS networks. Leica Geosystems in partnership with CRKennedy have
established Smartnet Aus (Smartnet Aus, 2009) predominantly around major capital
cities and in agricultural and mining regions. These networks are entirely separate from
the government owned and operated networks but in some instances could be used in
partnership – if the user has the correct equipment on their machine and have paid the
subscription fees.

South Australia is of particular interest for CORS users as Smartnet Aus is currently the
only option in the absence of any government supported CORS stations (see figure 6).

GPSnet Perth and SunPoz (Qld) are similar to SmartNet Aus however these networks
were established as partnerships between Trimble and the WA and QLD government
authorities respectively.
    Figure 6 – Smartnet Aus in the vicinity of South Australia (Smartnet Aus, 2009)

Precision Agriculture terminology

A few definitions are given below in order that the reader is clear on the later discussion
in this paper.

RTK GPS – A differential GPS technique whereby a so-called ‘base station’ and a
‘rover’ observe GPS signals simultaneously and the base communicates with the rover
(via radio or mobile phone) to allow high precision positioning at the rover which is fixed
onto agricultural machinery. Positions are good to 2 cm precision and updated every
second (or more frequently if required). One base station could service many compatible
rovers over distances of up to 10 – 15 kms depending on conditions. Also called single
base station RTK.

Network RTK – Provides RTK positions over longer distances by utilizing a network of
communicating CORS stations. CORS stations can be located 50 – 70 km apart and still
maintain 2 cm precision positions using GPRS or Next G mobile phone communications.

CMR+ - (Compact Measurement Record). A Trimble proprietary communications
protocol. The ‘language’ used to send data from a base station to a rover. Was developed
to overcome limitations in previous RTCM protocols and has become a quasi-industry
standard.
RTCM3.0 – (Radio Technical Commission for Martime communications). The RTCM is
a committee that develops communications protocols called RTCM version x.x. Previous
versions of RTCM (2.1, 2.2, 2.3) were quite cumbersome, hence the move to CMR+. In
recent times, the new RTCM version 3.0 is a much better protocol and should re-emerge
as the real industry standard.

NTrip – a protocol that enables streaming of DGPS, RTK or network RTK corrections
over the internet.

Community base stations – Base stations set up by local farmers or agricultural
companies to service local regions for precision agriculture applications.

Controlled Traffic Farming – GNSS guidance systems and automated steering
constrain tractor movements to permanent wheel tracks year after year, reducing soil
compaction producing higher yields and lower fuel costs.

Inter-row sowing – A technique whereby seeds are sown between rows of crops from
previous years to maximise crop yields.

OmniSTAR HP or XP and John Deere SF2 – Commercial products which give 10cm
accuracy positioning anywhere on the Earth using their own derived satellite corrections.
Often used for CTF farming.

Why precision agriculture?

Many studies have been conducted on the economic benefits of precision agriculture
(Allen Consulting Group 2008, Schofield et al, 2007). Precision agriculture is split into
two categories: Controlled Traffic Farming and Inter-row sowing. CTF techniques may
be suitable at up to the 10cm level whereas inter-row sowing generally requires 2cm
accuracy.

Benefits of Controlled Traffic Farming
Schofield et al, (2007) estimates that the 12% of a paddock that is sacrificed for wheel
tracks is more than compensated by gains in improved soil structure and water holding
capacity contributing to better root penetration and therefore higher yields. There is also
less crop damage and more targeted application of herbicides and fertilizers.

Lower powered tractors are sufficient when aligning along compacted wheel tracks
leading to reduced capital costs and ongoing fuel and maintenance costs not to mention
reduced greenhouse gas emissions. Significantly there is also less wastage from overlap
sprays. Another important benefit is the reduced stress to the farmer as physical steering
is automated and adopting these techniques leads to a reduced chance of a failed crop.
Benefits of Inter-row sowing
It has been shown that sowing new crops between existing crop rows reduces disease and
eliminates the need to clear the stubble of existing rows (McCallum, undated).
Additionally the stubble provides protection for the new growth, improves weed control,
offers higher levels of soil nutrient and improves soil water retention. Allens Consulting
Group (2008) report increases due to reduced input costs from herbicides and pesticides
of between 3 – 10% and increased wheat yields in South Australia using inter-row
sowing of 8%.

Advantages of CORS for precision farming

Pedersen & Yule (2008) describe a number of technical issues which users of GNSS for
precision farming need to be wary. Community base stations have been set up by farmers
or cooperating groups of farmers to service a local need – often in the surrounding 10kms
of the base station. The expense of the base station and the machine mounted rovers is
borne by the farmer(s). The position programmed into the base station is usually random
within a 5 m radius (even if averaged solutions are used) and cannot be reproduced unless
the exact coordinate is recorded for reuse. All steering and guidance operations therefore
relate to this base station and its adopted coordinate. In the event of an outage at the base
station (due to power loss, communications issues, lightning strike etc), if this coordinate
was not recorded and re-entered into the base station, then a new derivation of the base
station coordinate may be different by several metres. So the system will still deliver
positions to 2cm precision but they may be several metres different to the positions from
the previous base station coordinate. Of course the base station should not be physically
moved either as this will also change the coordinate.

If a farmer uses a neighbouring base station, the coordinate of this second base station
may not be aligned to the coordinate from the original base station by several metres,
meaning that either the farmer must use just one base station for their entire farm or
‘survey in’ the second base station with respect to the first for consistency. This is really
the task of a surveyor. The advantage of CORS networks with regard to community base
stations is that by default they have a fixed and repeatable base station coordinate in the
national coordinate reference frame (GDA94), they are less likely to have outages due to
power or communications and a user does not have to purchase the infrastructure –
although a subscription fee to use the service is required.

Pedersen & Yule (2008) report that many CTF users may only require 10cm accuracy
and therefore use the OmniSTAR or John Deere systems. These systems require some
initialization time after which a 10cm precision or so called ‘track to track accuracy’ is
achieved. This is a misnomer. Accuracy refers to an absolute position on the ground.
When the system is switched off and re-intialised, positions can shift by more than a
metre although the track to track accuracy thereafter remains precise to 10cm. This
problem will not be experienced when using a CORS network.

Additionally a collection of community base stations may not comprise the same brand of
GPS receiver and antenna which could compromise the accuracy of the resultant
positions on the agricultural machine and incompatible communications protocols
between different base stations and farm machinery might mean there is no positioning at
all. For instance one base station might broadcast CMR-w (a variant of CMR+) and a
user might require RTCM3.0. A standardized CORS network can provide a range of
services and protocols that a user can choose and use with confidence without any extra
expert knowledge – and the service is guaranteed.

Conversely a collection of community base stations may have been set up by a single
supplier to have consistent hardware and communications protocols. Users can simply
switch between base stations as they work in and out of range of neighbouring base
stations. This configuration also promotes sales of the same equipment for the supplier
who set up the community base stations. However these stations are not networked. They
deliver 2 cm accurate single base RTK results but many more stations are required for
coverage of the same area compared to CORS (note: SmartNet Aus is a private CORS
network).

A CORS network of sufficient density can produce network RTK solutions – positions
accurate to 2 cm in the GDA reference frame of the CORS network for distances up to 35
km from the nearest base station. An additional advantage of the network RTK approach
is the solution is often more robust as more than one station helps ensure correct
positioning and guards against a false initialization which is less recognisable with a
single base RTK scenario. Initialisation has been shown to be 30 – 50% faster using
network RTK techniques which can only be employed within a CORS network (DoL,
2009).

In recent years the GPS system is undergoing a major modernization phase whereby new
satellites and signals are coming into operation. Keeping pace with this is the re-
invigorated Russian Glonass system which will offer a similar number of satellites as
GPS by 2010. The European GNSS, Galileo and the Chinese Compass GNSS system will
be operational before 2015. Altogether more than 100 navigation satellites providing
positioning for a wide range of precision navigation applications, including agriculture
will soon be available. CORS network administrators will be the best placed authorities
to keep up with new developments and ensure their networks are up-to-date to supply
their users. Operators of community base stations will be hard pressed to stay on top of
the high pace of change.

In rural and regional Australia, the best way of accessing precise positioning services
from CORS networks is via the mobile phone network using the Ntrip protocol. The new
Next-G technology provides the bandwidth and is relatively cheap. Some manufacturers
are supplying CORS-ready equipment to be mounted on machinery equipped with a
Next-G modem, compatibility with a fully networked CORS solution and a built in
booster in the rover giving significantly improved mobile phone cover compared to
handsets. Also CORS network corrections can be sent over the internet and re-broadcast
over a farm using a radio link. There is little extra latency in the broadcast solution with
this configuration.
Conclusions

Current and future benefits due to precision GNSS for agriculture are estimated at $152 –
206 million in 2008 and projected to $1005 – 1357 million in 2030 assuming a
standardised national CORS network (Allen Consulting Group, 2008). By way of
example, the NSW government will spend $7.25 million over 5 years in capital
investment for CORSnet NSW (DoL, 2009). This paper argues that CORS networks are
the best way to ensure a robust precise positioning infrastructure.

However, farmers and agricultural companies cannot be blamed for setting up their own
collections of community base stations in the absence of an established CORS network in
their region. Nearly all cotton farmers who use GNSS techniques use their own
community base stations (Lorimer, 2008) as no CORS networks exist in these regions.
Indeed it is the demands of these important applications which are pushing state and
federal governments to accelerate the establishment of CORS networks.

Kondinin Group (2006) found that only 15% of broad acre wheat growers in Western and
South Australia were using GNSS technology that provide sub-10cm accuracy. One Ag
consulting firm estimated that only about 2% of farmers are presently doing inter-row
sowing (Nix, 2009, personal communication).

The pitfalls of community base stations used for CTF and inter-row sowing have been
detailed in this paper. The uptake of GNSS technology for agriculture is restricted by
these complications and the slow pace of CORS network establishment across Australia.

Precision agriculture is an important application of future CORS networks and
agricultural lobbyists should work closely with state and federal government land
authorities to ensure their regions are well covered by CORS positioning infrastructure.

Acknowledgements

The author would like to thank Martin Nix of Navonix consulting for valuable expert
discussions about the benefits of CORS for precision agriculture.

References

Allen Consulting Group (2008) Economic benefits of high resolution positioning
services, prepared for the Victorian Dept of Sustainability and Environment and the
Cooperative Research Centre for Spatial Information.

DoL (2009) Department of Lands, NSW, CORS network testing, Internal research report.
Kondinin Group (2006) Affordable Gear Sees Guidance Use Double, Research Report.

Lorimer, R. (2008) User Needs Analysis Project 1.04, Prepared by Position One
Consulting, April.

McCallum, M. (undated) Multiple Benefits from Inter Row Sowing with 2 cm RTK
GNSS, Report prepared by McCallum Agribusiness Consulting, South Australia.

Nix, M. (2009) Navonix Consulting

Pedersen, H. H. and Yule, D. (2008) European Network GPS and future directions, CTF
2008 Conference, Dubbo, Australia 12 – 14 August (www.actfa.net )

Rizos, C., 2008. The contribution of GNSS CORS infrastructure to the mission of
Modern Geodesy. 7th Int. Symp. & Exhibition on Geoinformation, Kuala Lumpur,
Malaysia, 13-15 October, CD-ROM procs.

Schofield, N., Chudleigh, P. and Simpson, S. (2007) ‘Case study 9: Controlled Traffic
Farming’ in Land and Water Australia’s Portfolio Return on Investment and Evaluation
Case Studies, March 2007.

Smartnet Aus (2009) http://smartnetaus.com/ [ Last accessed 20 July 2009 ]

								
To top