Mental Health Monitoring Mobile Robot

A mobile four-wheeled robot based on ESP32-CAM that collects environmental and facial data, computes mental health scores, visualizes them in real time with InfluxDB and Grafana, and sends alerts when scores drop below defined thresholds.

---

Features

• Environmental Sensing: Measures temperature, humidity (DHT11), illumination (photoresistor) and air quality (MQ135) via ESP8266 (simulated script).
• Facial Recognition: Captures frames on ESP32-CAM and sends them to a local face detection server, then to a face recognition server for per-person mental health scoring.
• Data Storage & Visualization: Pushes sensor and mental health data to InfluxDB and visualizes dashboards in Grafana (global and per-person views).
• Alerts: Sends email notifications when a person’s mental health score falls below configurable thresholds.
• Remote Control: Hosts a custom web interface on the ESP32-CAM for controlling robot movement and live video feed.

---

Hardware Components

• Robot Platform: 4-wheel chassis with motor drivers

• Camera & Controller: ESP32-CAM module

• Sensor Node: ESP8266 (currently simulated)

  • DHT11 (Temperature & Humidity)
  • Photoresistor (Light / Darkness)
  • MQ135 (Air Quality)

---

Software Stack

• Firmware:

  • ESP32-CAM Arduino sketch for camera, web server, and InfluxDB client
  • ESP8266 Arduino sketch (or Python simulation) for sensors

• Back-End Servers:

  • Local Face Detection Server
  • Local Face Recognition and Emotional Analysis Server

• Data Storage:

  • InfluxDB (time-series database)

• Visualization:

  • Grafana dashboards [global and per-ID panels, Weather Data (Temperature, Humidity, Air Quality, ...)]

• Alerts:

  • Python script using smtplib for threshold-based email notifications

---

Setup & Installation

Prerequisites

1. Hardware assembled robot chassis, ESP32-CAM, ESP8266 (or simulation environment).

2. Software installed on host machine:

   • Arduino IDE (ESP32 & ESP8266 cores)

   • InfluxDB 2.x

   • Grafana 8.x+

   • Python 3.8+ with dependencies:

     pip install opencv-python influxdb-client 

Steps

1. InfluxDB & Grafana:

   • Install and start InfluxDB. Create mental_health bucket and generate an API token.
   • Install and start Grafana. Add InfluxDB as a data source and import provided dashboard JSON (in dashboards/ folder).

2. Servers:

   • Launch face detection server:

     python face_detection_server.py

   • Launch recognition server:

     python face_recognition_server.py

3. Firmware:

   • Flash esp32cam/esp32cam.ino with your bucket URL, token, and thresholds.
   • Flash esp8266/sensors.ino (or run esp8266-simulator.py) to send sensor readings to InfluxDB.

4. Web UI:

   • ESP32-CAM hosts the control UI at http://<ESP32CAM_IP>/.
   • Use the UI to drive the robot and view live video.

---

Usage

1. Power on the robot and ensure Wi-Fi connectivity.

2. Access the web UI on ESP32-CAM to control motors and view live feed.

3. Monitor live dashboards in Grafana:

   • Global Dashboard: overall environmental and mental health trends.
   • Per-Person Dashboard: individual scores and conditions.

4. Email alerts are sent automatically when scores fall below configured thresholds.

---

Configuration

Edit :

• InfluxDB URL, bucket, organization, token
• Thresholds for warning_score and critical_score
• Email SMTP server settings and recipient lists

---

Simulated Sensor Data

If using simulation, run:

python esp8266-simulator.py 

This script will publish random but realistic DHT11, photoresistor, and MQ135 readings to InfluxDB every 10 seconds.

---