An embedded workspace monitor designed around an asynchronous C++ state machine running locally on an ESP32 microcontroller. The system actively reduces digital eye strain, static muscle fatigue, and cognitive environmental decline using research-backed thresholds from the American Optometric Association (AOA).
To ensure 100% video and data privacy, the entire ecosystem relies strictly on non-optical radar tracking arrays and edge processing. It operates completely independently of external cloud services or camera-based tracking networks.
---
Real-time telemetry — posture distance, thermal readings, lux analytics, pomodoro sync, wellness score, and live session audit logs.
Automatic focus sprint pause triggered when operator absence exceeds 10 seconds. Browser push notification issued simultaneously.
Hardware link configuration, OTA calibration sliders, timestamped system security log stream, and 7-day local privacy analytics summary.
---
The system separates its physical components into three functional layers, isolating low-current CMOS logic lines from high-current inductive mechanical loads.
| Hardware Layer | Component Model | ESP32 GPIO Pin | Interface Type | Electrical Isolation Routing |
|---|---|---|---|---|
| Perception (Inputs) | HC-SR04 Proximity Radar | Trigger: G18 / Echo: G19 |
Digital I/O | Direct Logic Rail |
| DHT11 Microclimate Unit | Data: G4 |
Single-Wire Bus | 10kΩ Pull-Up Resistor | |
| LDR Photoresistor Array | Analog: G34 |
ADC Register | Voltage Divider Network | |
| Actuation (Outputs) | SG90 Status Needle Servo | PWM: G26 |
Hardware Timer | Isolated External 5V Rail |
| Coin Haptic Vibration Motor | Control: G27 |
Digital Out | ULN2003 Darlington Driver | |
| Active Alarm Siren | Trigger: G25 |
Digital Out | ULN2003 Darlington Driver | |
| Passive Piezo Chime Buzzer | Tone: G32 |
LEDC PWM Channel | Direct Driver Channel |
---
The background loop() coordinates four automated safety modules without running blocking code delays:
Manages an isolated 25-minute work sprint and 5-minute break sequence. The frontend telemetry loop queries the on-chip HTTP server at /readSensors every 1000ms. If a user stands up or completely leaves the radar sweet spot for more than 10 continuous seconds during a focus sprint, it triggers an automation breakdown, records a distraction event, drops the mode to idle, and sounds a warning tone.
- AOA Proximity Guard: Detects if a user slouches closer than 50 cm. It grants a 3-second buffer window before pulsing a tactile vibration reminder.
- Static Stiff Monitor: Prolonged fixed sitting creates deep fatigue. If body movement shifts by less than 3 cm over 20 continuous minutes, the mechanical status needle swings to its 45° break zone and delivers a firm 1-second physical shake.
Productivity declines rapidly above 25°C, with severe cognitive accuracy degradation at 28°C+.
- 25°C to 29°C: Sounds a polite double-chirp tone via the piezo channel once every 5 minutes as a microclimate warning.
- ≥ 30°C: Triggers an emergency system override, swinging the status servo to 180° and latching the high-output active siren.
To avoid high-contrast computer vision strain (working on a bright screen in a pitch-black room), the logic checks the LDR sensor. If focus mode is running and ambient illuminance drops below a raw ADC value of 1250, a non-blocking timer triggers a three-tone descending alert chime once every 5 minutes.
---
-
Clone this repository to your local directory.
-
Open
firmware/workspace\_fortress.inowithin the Arduino IDE. -
Update the global placeholder network credential strings (
ssidandpassword) with your local access point details. -
Install the required libraries via Arduino Library Manager:
DHT sensor libraryby AdafruitESP32Servoby Kevin Harrington
-
Compile and flash the sketch to your ESP32 board node.
-
Launch your browser and navigate to the local IP printed on your IDE Serial Monitor at
115200baud.
---
| Component | Model | Qty |
|---|---|---|
| Microcontroller | ESP32 DevKit V1 | 1 |
| Proximity Radar | HC-SR04 Ultrasonic | 1 |
| Microclimate Sensor | DHT11 | 1 |
| Light Sensor | LDR Photoresistor | 1 |
| Status Actuator | SG90 Micro Servo | 1 |
| Haptic Feedback | Coin Vibration Motor | 1 |
| Alarm Output | Active Buzzer 5V | 1 |
| Chime Output | Passive Piezo Buzzer | 1 |
| Motor Driver | ULN2003 Darlington Array | 1 |
| Display | 1602A LCD (16-pin parallel) | 1 |
| Power Regulation | LM2596 Buck Converter | 1 |
| Relay | 5V Single Channel Relay | 1 |
---
| Layer | Tool |
|---|---|
| Microcontroller Firmware | C++ / Arduino IDE |
| Web Dashboard UI | Antigravity AI (Vibe Coding) |
| Backend API | Flask (Python) |
| Hardware Protocol | ESP32 WebServer (HTTP/JSON) |
| Sensor Libraries | DHT, ESP32Servo |
All processing occurs entirely on-device. No camera, no cloud API, no data exfiltration. The 7-day analytics summary is stored in browser localStorage — encrypted and confined to the local machine. The ESP32 HTTP server is LAN-bound and never exposed to the public internet.
---
| Milestone | Status |
|---|---|
| Firmware state machine | Complete |
| Flask backend integration | Complete |
| Web dashboard (HTML/JS) | Complete |
| OTA calibration via sliders | Complete |
| Session log audit stream | Complete |
| Hardware assembly (breadboard) | Complete |
---
ASWIN MENON
Aswin Menon — ECE, VIT Vellore
Embedded Systems | Edge AI | IoT
GitHub