Real-Time Agricultural Environmental Monitor

Problem Statement

Storage facilities for potatoes require precise environmental control. Temperature above 50°F accelerates sprouting, while high humidity causes rot. Manual monitoring misses critical changes that happen overnight.

Solution Architecture

Built a dual-interface monitoring system:

• Local: 1.5” OLED display for at-a-glance readings
• Remote: Mobile-responsive web dashboard
• Integration: JSON REST API for third-party systems

Technical Implementation

Hardware Configuration

• MCU: ESP32-WROOM-32
• Sensor: DHT22 (±0.5°C, ±2% RH accuracy)
• Display: SSD1306 128x64 OLED
• Power: USB with battery backup option

Non-Blocking Event Loop

Key to achieving less than 50ms response time:

void loop() {
    // Non-blocking sensor read
    if (millis() - lastSensorRead > SENSOR_INTERVAL) {
        readSensorAsync();
        lastSensorRead = millis();
    }
    
    // Handle web requests without blocking
    server.handleClient();
    
    // Update display without blocking
    if (millis() - lastDisplayUpdate > DISPLAY_INTERVAL) {
        updateDisplayAsync();
        lastDisplayUpdate = millis();
    }
    
    // Check alerts
    checkThresholds();
}

Web Interface

Mobile-responsive dashboard with real-time updates:

// Auto-refresh data every 5 seconds
setInterval(async () => {
    const response = await fetch('/api/sensor');
    const data = await response.json();
    updateDisplay(data);
}, 5000);

API Endpoints

RESTful API for integration:

• GET /api/sensor - Current readings
• GET /api/history - 24-hour history
• GET /api/stats - Min/max/average
• POST /api/alerts - Configure thresholds

OLED Burn-in Prevention

Implemented 4-phase pixel shifting:

void shiftDisplay() {
    static uint8_t phase = 0;
    int xOffset = (phase % 2) * 2;
    int yOffset = (phase / 2) * 2;
    display.setCursor(xOffset, yOffset);
    phase = (phase + 1) % 4;
}

Features

✅ Dual Interface: Local OLED + Web Dashboard
✅ Real-time Monitoring: Less than 50ms response time
✅ Historical Data: 24-hour rolling buffer
✅ Alert System: Configurable thresholds
✅ NTP Time Sync: Accurate timestamps
✅ Auto-Recovery: Reconnects after WiFi loss
✅ JSON API: Easy integration

Performance Metrics

• Response Time: 47ms average
• Accuracy: ±0.5°C, ±2% RH
• Uptime: 99.8% over 30 days
• Memory Usage: 42% RAM, 67% Flash
• Power Consumption: 0.8W average

Real-World Impact

During testing, the system detected a humidity spike at 3 AM caused by a failed ventilation fan. The alert allowed maintenance to fix the issue before condensation formed, preventing potential loss of 500 lbs of stored potatoes.

Future Enhancements

• LoRaWAN for extended range
• Multiple sensor support
• Machine learning for predictive alerts
• Solar power option
• Mobile app with push notifications

Resources

• Full Source Code