AgTech-Precision Agriculture: Lecture 1

What is Precision Agriculture?

• Technically: Observation, impact assessment and timely strategic response to fine-scale variation in causative components of an agricultural production process.

• Practically: Doing the right thing at the right time, in the right place, to the right animal.

• Core idea: Use data to manage variability and optimize inputs and outputs.

Why use Precision Agriculture (PA)?

• Driving goals: feed the world, efficient/quality/safe food, environmental protection, rising knowledge demand.

• Key motto: If we can measure it, we can manage it.

Variability in PA

• Types of variability:

  • Temporal variability: variation over time (e.g., rainfall, yield over years).

  • Spatial variability: variation across an area (e.g., soil pH, biomass).

  • Individual variability: variation among individuals (e.g., animal weight gain).

• PA targets fine-scale variability to apply management where needed.

PA as a process

• Step 1: Observation & data collection

• Step 2: Data interpretation & evaluation

• Step 3: Implementation of a management plan

Paddocks as the management unit

• Paddocks are the basic units for decision making and management in PA.

Goals of PA in practice

• Manage variability to improve yield and resource use efficiency.

• Move from uniform to site-specific management.

PA terminology: site-specific and individual management

• Soils/plants: Site-specific management

• Animals: Individual animal management

GNSS and PA overview

• GNSS: Global Navigation Satellite Systems provide positioning data for PA tasks (mapping, guidance, variable-rate application, etc.).

GNSS: What’s out there

• Global systems: GPS (US), GLONASS (Russia), Galileo (EU), Beidou (China), NAVIC (India), QZSS (Japan).

• Constellations vary in number of satellites and coverage.

How GNSS works (concept)

• Each satellite has a known location and transmits signals.

• Ground receiver measures distance (range) to satellites from signal travel time.

• With multiple satellites, the receiver triangulates its position.

• Basic relation: d = v t where d is distance, v is signal velocity, and t is time of flight.

GNSS basics: requirements

• Accurate time signals from satellites and a receiver clock.

• A constellation of accurate clocks and a ground receiver clock.

• Positioning relies on solving distance to multiple satellites.

GNSS: instantaneous time example

• At an instant (e.g., 4.10), signals from several satellites are used to compute receiver position.

• The satellites’ orbits are precisely known to enable this solution.

GNSS accuracy concepts

• Minimum satellites for stationary point: 4

• Minimum satellites for moving position: 5

• Positioning accuracy depends on geometry and corrections.

Error sources in GNSS

• Satellite clocks and orbit errors

• Ionospheric and tropospheric delays

• Receiver environment (multipath)

• Satellite constellation geometry (GDOP)

Reducing GNSS errors

• Use good GDOP values (better geometry).

• Average multiple measurements.

• Differential readings (corrections).

• Time-of-flight modelling (advanced).

Geometric Dilution of Precision (GDOP)

• GDOP measures how satellite geometry affects positioning error.

• Lower GDOP = better potential accuracy; poor GDOP amplifies errors.

GDOP visuals (concepts)

• Good GDOP: wide, well-distributed satellites over the sky.

• Poor GDOP: satellites clustered together or blocked view.

Multi-path and accuracy considerations

• Multipath and visual obstruction can worsen accuracy even with good GDOP.

• Averaging and correction methods help mitigate these effects.

Averaging GNSS locations

• For uncorrected GNSS, averaging can reduce random error to roughly ±5–10 m.

Differential GNSS (DGNSS)

• Concept: use two receivers - a base at a known location and a rover at unknown location.

• The base’s known error is used to correct the rover’s readings.

• Accuracy improves from ~10 m (uncorrected) toward 1–5 m or better depending on setup.

How DGNSS works

• Base station records true position and broadcasts corrections to rovers.

• Corrections can be transmitted in real time or post-processed later.

• Correction signals are sent via radio, cellular, or satellite link.

Real-time vs post-processing corrections

• Real-time corrections provide immediate improvements (RTK, DGPS).

• Post-processing applies corrections after data collection (common in some workflows).

Typical DGNSS configurations

• RTK: local base + rover with radio or network link; cm to decimeter accuracy.

• CORS / NRTK: regional corrections via network; decimeter to centimeter accuracy.

• SBAS / PPP: wide-area corrections via satellite or network; decimeter to tens of cm accuracy depending on conditions.

GNSS accuracy overview (typical horizontal)

• Autonomous / uncorrected: ~10 m

• DGNSS: ~1–5 m

• SBAS: ~1 m

• RTK / CORS / PPP: ~2–10 cm

• Vertical accuracy: typically ~2× horizontal accuracy

GNSS usage in PA: practical takeaways

• Positioning enables controlled traffic farming, precise input placement, yield mapping, and spatially variable management.

• On-the-go sensing (e.g., yield monitors, soil sensors) benefits from real-time corrections for accuracy.

• Targeted applications: variable-rate application, inter-row sowing, contour paddocks, and decision support through GIS/maps.

Summary concepts for PA & GNSS

• PA = manage variability to increase production and reduce environmental impact.

• Variability types: temporal, spatial, individual.

• PA process: observe, interpret, implement.

• GNSS provides the location data essential for site-specific decisions.

• Accuracy depends on satellite geometry (GDOP), corrections, and environment.

• Real-time corrections (RTK/DGNSS) offer the highest practical accuracy for field work.