IoT Solution

FuelShield

Stopping fuel theft before it happens.
Real-time telemetry, anomaly detection, and fleet intelligence.

Impact

20%

Reduction in fuel expenses. Tested across 50+ commercial fleet vehicles, improving accountability and transparency.

99%
Ultrasonic Precision
Live
GPS & Geofencing
Next.js
Firebase
MQTT
ESP32
GSM
Recharts

The Problem

Fuel theft is one of the most silent yet expensive problems in commercial fleets. Traditional fuel logs and manual audits fail to detect real-time losses, leading to revenue leakage, disputes, and poor accountability.

The Solution

FuelShield was built to make fuel behavior visible, measurable, and actionable. It continuously monitors fuel levels using high-precision ultrasonic sensors and instantly alerts fleet owners of any anomalies.

Key Capabilities

Real-Time Monitoring

Ultrasonic sensors track fuel levels continuously with noise filtering for accurate readings.

GPS & Geofencing

Detects theft independent of movement and alerts if fuel is accessed outside designated zones.

Instant Theft Alerts

MQTT-based low-latency delivery of sudden drop alerts with severity classification.

Fleet Analytics

Visual dashboard for fuel usage trends, historical theft records, and vehicle performance.

Secure Architecture

Firebase Authentication and cloud-hosted real-time database for multi-vehicle support.

Edge Processing

ESP32 handles sensor polling and anomaly detection locally before transmission.

Data from the Tank.

FuelShield bridges embedded hardware and modern web systems. An **ESP32-based edge device** polls ultrasonic sensors at high frequency, filtering noise before calculating the rate of fuel change.

When abnormal drops are detected, the device publishes an **MQTT event** containing fuel data and GPS coordinates. The backend consumes these events to update the dashboard and trigger alerts in sub-second time.

Hardware Layer

ESP32 + Ultrasonic Sensor + GSM Fallback.

Communication

MQTT (HiveMQ) for lightweight pub/sub.

Application

Next.js Dashboard + Firebase Realtime DB.

mqtt_client.ts
const client = mqtt.connect(BROKER_URL);

// Listen for sensor anomalies
client.on('message', (topic, message) => {
const data = JSON.parse(message.toString());

// Check for sudden fuel drop
if (data.drop_rate > THRESHOLD) {
// Trigger critical alert
sendAlert({
truckId: data.id,
location: data.gps,
severity: 'CRITICAL'
});
}
});
Back to HomePart of the 2024 Portfolio