DeskBot AI Companion (ESP32)

A fully autonomous AI desktop companion robot powered by ESP32 and a Mobile Node.js Brain.

🌟 Features

• Smooth Eye Engine: Custom OLED animation engine (squircle eyes).
• Mobile-First Design: Connects automatically to a Phone Hotspot.
• Termux Brain: Offloads AI processing to an Android phone running Node.js.
• Crash-Proof: Optimized power management to prevent brownouts on USB power.
• Offline Mode: Continues to interact (Touch/Motion) even without WiFi.

🛠️ Hardware & Effective Usage

• ESP32 (Dev Module): Main controller with WiFi/Hotspot connection, WebSocket FreeRTOS queue, and brownout protection.
• SH1106 OLED Display (1.3"): Renders dynamic squircle eyes, blinking animations, emotional expressions, and stopwatch.
• SG90 Servo x1: Non-blocking 50Hz servo controller for gestures (shake, tilt, droop, jolt) and idle eye tracking.
• Two jumperwire directly from esp32 for touch sensing: Direct capacitive touch sensing on GPIO 4 (Head/T0) and GPIO 15 (Side/T3) with 2-sample noise debouncing.
• Speaker & I2S Amp: Audio output via I2S DMA with time-multiplexing for sound effects and TTS streaming.
• Buzzer: Non-blocking tone sequences for startup, emotions, sleep, and DS3231 alarm ringing.
• ldr sensor module: Ambient light detection for automatic dark sleep mode and UI reporting.
• DS3231 Real Time Clock Memory Module: Hardware I2C RTC with automatic server time syncing, room temperature monitoring, and hardware alarm ringing.
• PIR Motion Detection Sensor: Presence detection with crowd-proof hold timers and automatic waking.
• Ultrasonic Sensor: Distance measurement (Trig/Echo) with timeout filtering and proximity zoning (Near/Far).
• 16Bit RGB LED (WS2812B): 16-LED NeoPixel ring with 7 animation modes, voice reactivity, and mood colors.
• Smartelex i2s mems microphone: 16-bit DMA microphone with volume-based VAD and audio chunk streaming over WebSockets for speech recognition.

AI Services Module for DeskBot

Supports:

• • Edge TTS (FREE, works everywhere - PC & Phone!)
• • Piper TTS (local, free - PC only)
• • Ollama LLM (local, free), GROQ API OR Gemini API

TTS Priority:

• Edge TTS (default - FREE, online, great quality)
• Piper TTS (if installed locally)
• Browser TTS (fallback)

Configuration

Before starting the server, you must configure your local network and API credentials.

• Open your src/config file

• Update the Wi-Fi Name and password to match your mobile hotspot.

• Also update you inet address to WS_HOST (You can find you inet address by running ifconfig in termux)

• Open server/public/ai-services.js

• Insert your API key into the required field.

🚀 How to Run (Termux)

1. Install Termux on Android.
2. Run pkg install nodejs
3. Run npm -v to check npm
4. npm install express
5. Clone this repo.
6. Go to server folder cd server
7. Run node server.js
8. Turn on Hotspot and power up the robot.