3  Capstone Project 2: Smart Agriculture System

One of three capstone project briefs in this series – along with Capstone Project 1: Smart Environment Monitor and Capstone Project 3: Fleet Tracking System. The shared capstone approach (reviewable claims, the thin-slice method, and interface evidence) is introduced in Part 1; this part works through the Smart Agriculture System brief.

The second brief adds actuation. That changes the review question from “did the system observe correctly?” to “did the system act safely when soil, weather, water level, and schedule evidence disagreed?”

3.1 Capstone Project 2: Smart Agriculture System

Project Overview

Domain: Precision Agriculture / Smart Farming

Difficulty: Advanced

Duration: 6-8 weeks

Team Size: 2-3 people

3.1.1 Project Description

Build an automated plant care system that monitors soil conditions, environmental factors, and plant health, then automatically waters plants based on intelligent algorithms. The system should work both indoors (houseplants) and outdoors (garden), with solar power capability for remote deployment.

3.1.2 Project Context

Smart irrigation projects show how IoT can reduce manual monitoring, improve watering consistency, and make resource use visible. Your system should demonstrate closed-loop sensing and actuation while still protecting plants from unsafe automated decisions.

3.1.3 Requirements Specification

3.1.3.1 Functional Requirements

ID	Requirement	Priority
F1	Measure soil moisture at multiple depths	Must Have
F2	Measure soil temperature	Must Have
F3	Measure ambient temperature and humidity	Must Have
F4	Control water pump/valve	Must Have
F5	Implement watering schedules	Must Have
F6	Smart watering based on soil moisture	Must Have
F7	Measure light levels (PAR for plants)	Should Have
F8	Detect water tank level	Should Have
F9	Weather forecast integration	Should Have
F10	Multi-zone support (different plants)	Could Have
F11	Camera for visual monitoring	Could Have
F12	Plant health ML classification	Could Have

3.1.3.2 Non-Functional Requirements

ID	Requirement	Target
NF1	Solar-powered operation	3+ days without sun
NF2	Watering accuracy	±10% of target volume
NF3	LPWAN range	> 500m (LoRa)
NF4	Water savings vs. manual	> 30%
NF5	System cost	< $150

3.1.4 Recommended Hardware

Component	Options	Est. Cost
Microcontroller	ESP32, Arduino + LoRa	$10-20
Soil Moisture	Capacitive (not resistive!)	$3-8 each
Soil Temperature	DS18B20 waterproof	$3-5
Water Pump	12V submersible, 3-5L/min	$8-15
Relay Module	1-4 channel, 5V/12V	$3-5
Solar Panel	6V 3W or 12V 5W	$10-20
Battery	18650 Li-ion with BMS	$10-15
Water Sensor	HC-SR04 ultrasonic	$2-5
Enclosure	IP65 waterproof box	$10-15

Putting Numbers to It

Solar power sizing requires matching daily energy consumption with panel output. ESP32 + LoRa transmits once/hour (30s active at 200mA, rest at 5mA sleep).

Daily energy budget: \[E_{daily} = 24 \times \left(200mA \times \frac{30s}{3600s} + 5mA \times \frac{3570s}{3600s}\right) \times 3.3V\] \[E_{daily} = 24 \times (1.67 + 4.95) \times 3.3 = 523 \text{ mWh/day}\]

Solar panel output (5W at 5 hours equivalent sun): \[E_{solar} = 5W \times 5h = 25 \text{ Wh/day} = 25000 \text{ mWh/day}\]

Panel produces 48x more than needed. With 80% charging efficiency, we get 20Wh/day, enough to run the system and charge a 18650 battery with about 11Wh capacity in roughly half a day, leaving margin for cloudy days.

viewof esp_current = Inputs.range([50, 300], {value: 200, step: 10, label: "ESP32+LoRa Active Current (mA)"})
viewof esp_sleep = Inputs.range([1, 20], {value: 5, step: 1, label: "Sleep Current (mA)"})
viewof active_duration = Inputs.range([10, 120], {value: 30, step: 10, label: "Active Duration per Hour (seconds)"})
viewof solar_panel_watts = Inputs.range([1, 10], {value: 5, step: 1, label: "Solar Panel Power (W)"})
viewof sun_hours = Inputs.range([2, 8], {value: 5, step: 0.5, label: "Equivalent Sun Hours/Day"})
viewof voltage = Inputs.range([3, 5], {value: 3.3, step: 0.1, label: "Operating Voltage (V)"})

solar_calc = {
  const active_fraction = active_duration / 3600;
  const sleep_fraction = 1 - active_fraction;
  const hourly_energy_ma = esp_current * active_fraction + esp_sleep * sleep_fraction;
  const daily_energy_mwh = 24 * hourly_energy_ma * voltage;
  const solar_energy_mwh = solar_panel_watts * sun_hours * 1000;
  const ratio = solar_energy_mwh / daily_energy_mwh;
  const efficiency = 0.8;
  const net_solar = solar_energy_mwh * efficiency;
  return {daily_energy_mwh, solar_energy_mwh, ratio, net_solar};
}

html`<div style="background: var(--bs-light, #f8f9fa); padding: 1rem; border-radius: 8px; border-left: 4px solid #E67E22; margin-top: 0.5rem;">
<p><strong>Daily Energy Consumption:</strong> ${solar_calc.daily_energy_mwh.toFixed(0)} mWh/day</p>
<p><strong>Solar Panel Output:</strong> ${solar_calc.solar_energy_mwh.toFixed(0)} mWh/day (gross)</p>
<p><strong>Net Solar (80% efficiency):</strong> ${solar_calc.net_solar.toFixed(0)} mWh/day</p>
<p><strong>Solar/Consumption Ratio:</strong> ${solar_calc.ratio.toFixed(1)}x (${solar_calc.ratio >= 2 ? 'excellent margin' : solar_calc.ratio >= 1.2 ? 'adequate margin' : 'insufficient for cloudy days'})</p>
<p style="margin-top: 0.5rem; color: ${solar_calc.ratio >= 1.2 ? '#16A085' : '#E67E22'};">
${solar_calc.ratio >= 1.2 ? 'Status: solar sizing adequate for multi-day autonomy' : 'Status: increase panel size or reduce duty cycle'}
</p>
</div>`

Knowledge Check: Soil Moisture Sensor Selection

3.1.5 System Architecture

Layer	Responsibilities	Typical Components
Field node	Sense soil and weather conditions, decide whether local actuation is safe, and drive the pump or valve through a relay.	ESP32, soil moisture sensors, soil temperature sensor, pump relay, battery, charge controller
Gateway	Bridge long-range field traffic to the internet when Wi-Fi is not available at the planting area.	LoRa receiver, Raspberry Pi or ESP32 gateway, Wi-Fi or Ethernet uplink
Cloud services	Store readings, combine weather forecasts with local readings, and generate commands or alerts.	Weather API, decision engine, database, dashboard, notification service
Operator interface	Let the learner monitor moisture history, override watering, and inspect system health.	Web dashboard, alert rules, device status panel

Data and command flow:

1. Field node samples soil moisture, soil temperature, water level, and ambient conditions.
2. Node publishes readings over LoRa or Wi-Fi.
3. Gateway or cloud service stores readings and evaluates watering rules.
4. Safe watering commands return to the node, where local firmware checks final safety constraints before actuating the pump.

3.1.6 Smart Watering Algorithm

def should_water(soil_moisture, weather_forecast, last_watered):
    """
    Intelligent watering decision algorithm.

    Args:
        soil_moisture: Current soil moisture (0-100%)
        weather_forecast: Dict with rain_probability, temp
        last_watered: Datetime of last watering

    Returns:
        (should_water: bool, duration_seconds: int)
    """

    # Define thresholds (plant-specific)
    DRY_THRESHOLD = 30      # Water if below this
    WET_THRESHOLD = 70      # Never water above this
    RAIN_THRESHOLD = 60     # Don't water if rain likely
    MIN_INTERVAL_HOURS = 6  # Minimum time between watering

    # Check if too soon since last watering
    hours_since = (datetime.now() - last_watered).total_seconds() / 3600
    if hours_since < MIN_INTERVAL_HOURS:
        return (False, 0)

    # Check if rain is expected
    if weather_forecast['rain_probability'] > RAIN_THRESHOLD:
        return (False, 0)  # Let nature do the work

    # Check soil moisture
    if soil_moisture > WET_THRESHOLD:
        return (False, 0)  # Already wet enough

    if soil_moisture < DRY_THRESHOLD:
        # Calculate watering duration based on how dry
        deficit = DRY_THRESHOLD - soil_moisture
        duration = min(deficit * 2, 60)  # Max 60 seconds
        return (True, duration)

    return (False, 0)

Knowledge Check: Smart Watering Algorithm

3.1.7 Implementation Milestones

3.1.7.1 Week 1-2: Sensor Integration

• Soil moisture sensor calibration (dry/water/soil)
• Temperature sensor integration
• Basic data logging to serial

3.1.7.2 Week 3-4: Actuation & Control

• Pump/valve control with relay
• Basic watering schedule
• Water level monitoring

3.1.7.3 Week 5-6: Connectivity & Power

• LoRa or Wi-Fi connectivity
• Solar power system
• Battery management

Knowledge Check: Connectivity Protocol Selection

3.1.7.4 Week 7-8: Intelligence & Dashboard

• Weather API integration
• Smart watering algorithm
• Dashboard with plant profiles
• Documentation and testing

3.1.8 Evaluation Rubric

Criteria	Points	Description
Sensor Integration	20	Accurate, calibrated, reliable
Watering Control	20	Precise, safe, effective
Power Management	15	Solar-capable, efficient
Connectivity	15	Long-range, reliable
Algorithm	15	Smart decisions, water savings
Documentation	15	Complete, reproducible
Total	100

Checkpoint: Smart Agriculture

You now know:

• The must-have path measures soil moisture, soil temperature, ambient temperature and humidity, schedule state, smart watering, and pump or valve control.
• The safety logic includes dry and wet thresholds, a rain threshold, and a 6-hour minimum interval so one bad reading cannot drive repeated watering.
• The evidence package should cover calibration, relay behavior, LoRa or Wi-Fi choice, solar margin for 3+ days without sun, and outdoor credential protection.

Knowledge Check: Security for Outdoor Deployments

Worked Example: Budget-Constrained Capstone - Smart Agriculture on $120

Scenario: Student team with $120 budget wants to build a complete smart agriculture system (sensors + actuators + connectivity + cloud + dashboard). How to maximize functionality within budget?

Initial wishlist (exceeds budget): - ESP32 ($8) + LoRa module ($15) = $23 - 3x soil moisture sensors ($8 each) = $24 - Soil temperature sensor ($5) - Water pump + relay ($12) - Solar panel 5W ($18) + battery + BMS ($15) = $33 - Total: $97 … still need enclosure, wiring, cloud costs

Budget optimization:

Component	Expensive Option	Budget Alternative	Savings
Connectivity	LoRa module $15	Wi-Fi only (use ESP32 built-in)	$15
Soil sensors	3x capacitive $24	2x capacitive $16	$8
Solar system	5W panel + BMS $33	USB power bank ($12) + charge at night	$21
Enclosure	IP65 waterproof $15	Plastic food container + silicone ($4)	$11
Cloud	AWS IoT $5/mo	ThingsBoard free tier	$5/mo

Final BOM ($98):

• ESP32 development board: $8
• 2x capacitive soil moisture sensors: $16
• DS18B20 waterproof temperature sensor: $5
• 12V water pump: $8
• 5V relay module: $3
• 10,000mAh USB power bank: $12
• DHT22 (air temp/humidity): $6
• Jumper wires, connectors: $8
• DIY enclosure (food container + silicone): $4
• Tubing, fittings: $8
• Micro-USB cable: $3
• Breadboard for prototyping: $5
• Total: $98 (under budget!)

Trade-offs made:

1. Wi-Fi instead of LoRa: Requires router within 50m, but saves $15
2. 2 zones instead of 3: Still demonstrates multi-zone concept
3. USB power bank instead of solar: Manual charging, but reliable
4. DIY enclosure: Not IP65 rated, but adequate for protected outdoor location

Project scope adjusted:

Included in the first working version:

• Soil moisture monitoring for two zones
• Threshold-based automated watering
• Environmental monitoring for temperature and humidity
• Cloud dashboard with historical data
• Manual override through a web interface

Deferred until the core version is stable:

• Solar power; use USB charging first
• LoRaWAN; use Wi-Fi first
• Weather API integration; keep as a later enhancement

Key insight: Budget constraints force prioritization. The team identified must-have features for monitoring, control, and cloud visibility, then deferred solar, LoRa, and weather integration. The result is a functional system within budget.

3.2 Within This Series

Direction	Chapter	Topic
Previous	Capstone Project 1: Smart Environment Monitor	Indoor air-quality and comfort monitoring
Current	Capstone Project 2: Smart Agriculture System	Soil sensing and automated watering
Next	Capstone Project 3: Fleet Tracking System	GPS, cellular, and geofencing