1. Problem Identification
Cyclists often neglect minor tire pressure drops, which gradually deteriorate ride performance and can lead to failure or safety risk. The goal was to engineer a system that continuously monitors tire pressure in real time, detects abnormal behavior using automated anomaly detection, and alerts the rider instantly via Telegram.
2. Sensor Interface – Pressure Sensing Mechanism
The system uses the Honeywell ABPDANN150PGAA5 pressure sensor, which provides a precision analog voltage output proportional to pressure (0–150 PSI). The ESP32-C6 reads this voltage via its internal ADC and converts it into real-time pressure values.
The sensor is temperature-compensated and calibrated by design. For improved accuracy, multiple ADC readings are averaged before conversion into PSI.
3. Hardware Integration
The pressure sensor was connected to the bicycle tire valve using a coupling and tubing. The ESP32-C6 microcontroller was powered through a 18650 rechargeable Li-Ion battery.
To stabilize data before cloud integration, serial monitoring was used to validate the reading flow from raw ADC value → voltage → pressure.
4. Wireless Communication
After pressure processing, the ESP32 transmits the PSI data every 2 seconds over Wi-Fi using HTTP POST requests to a locally hosted Flask backend API.
The communication is formatted as a JSON payload: { "pressure": <value> }.
5. Backend Model & AI Anomaly Detection
The backend maintains a sliding window of recent readings and calculates:
- Rolling average pressure
- Deviation magnitude
- AI anomaly score based on deviation thresholding
If the pressure deviates substantially, it is marked as an anomaly.
“Backend CSV log showing live pressure data, deviation tracking, and anomaly score”
6. Real-Time Dashboard
A web-based dashboard was built using HTML and Plotly.js to visualize:
- Current tire pressure
- Average pressure over time
- Live pressure vs time graph
- Anomaly status and confidence value
- Total samples recorder
- "Smart Tire Monitor dashboard displaying real-time pressure and anomaly analytics”
7. Automated Alert via Telegram
When an anomaly is detected, the system immediately sends a notification to the cyclist via Telegram Bot API. The alert includes:
- Current pressure value (PSI)
- Anomaly score
- Timestamp
- Recommended action
“Instant Telegram push notification alerting user of abnormal tire pressure event”
8. Data Logging & Analytics
All the sensor readings and anomaly metrics are continuously logged as a CSV file on the backend. Users can download the logs via the dashboard for further diagnostics or performance analysis.
Video of the working:
Digikey BOM List
I am adding this list as prescribed by the rules in the event: LINK BELOW
https://www.digikey.in/en/mylists/list/CFE4GLSN29
Outcome
The system accurately detects tire pressure abnormalities in real time and informs the rider immediately, enabling proactive intervention before performance or safety is compromised. The integration of precision sensing, ADC-based signal processing, anomaly analytics, dashboard monitoring, and Telegram-based alerting demonstrates a complete end-to-end engineering solution.
Future Enhancements
- Cloud-based data storage and mobile app integration
- Automatic tire inflation using motorized pump
- Battery health monitoring
- Pressure-terrain detection using IMU feedback
- Multi-wheel multi-sensor scalability



