📱🤖 CellBot

Turn your old smartphone into a high-fidelity autonomous robot brain.

🚀 The Concept

CellBot challenges the idea that you need expensive hardware (Intel RealSense, LiDAR, Jetson Nano) to build a capable autonomous robot.

Instead, it leverages the Android smartphone you already own as a high-performance Sensor Hub. The phone captures 1080p video, fused IMU data, and GPS location, streaming it all via ultra-low latency WebRTC to a PC "Brain." The PC runs heavy Computer Vision and Path Planning algorithms, sending motor commands back to the phone to drive the chassis via a simple USB-OTG Arduino connection.

The result? A sophisticated autonomous platform for the price of a toy car.

---

✨ Key Features

• Split-Compute Architecture: Offloads heavy AI/CV processing to a PC while keeping the robot hardware lightweight.
• WebRTC Streaming: Ultra-low latency (<200ms) video and bi-directional JSON data over local Wi-Fi.
• Sensor Fusion: Combines Visual Odometry (Camera), GPS, and IMU data using an Extended Kalman Filter (EKF) for robust state estimation.
• Autonomous Navigation: Waypoint following, lane keeping, and PID-based steering control.
• Live Dashboard: Real-time web interface (Flask/Socket.IO) to monitor video, telemetry, and control the robot manually or autonomously.

---

🛠️ System Architecture

[ PC CONTROL CENTER ]                     [ ANDROID PHONE ]                       [ ARDUINO ]
   (Python/AsyncIO)                          (Kotlin)                             (C++)
        |                                       |                                     |
   [Main Loop] <---(WebRTC Video/Data)---> [Sensor Hub]                           [Motor Driver]
        |                                       |                                     |
   +----+----+                                  +---+---+                             |
   |         |                                  |       |                             |
[Perception] [Fusion]                        [Camera] [IMU/GPS]                       |
   |         |                                          |                             |
[Navigation] [Control]                                  +----(USB Serial OTG)----> [Parser]
   |         |                                                                        |
   +----+----+                                                                     [Motors]
        |
   [Dashboard] (Flask/SocketIO)

📦 Hardware RequirementsRobot Chassis:

2WD or 4WD kit.Motors: 4x DC Motors (TT Gearbox style recommended). Motor Driver: L298N or HW-130 Motor Shield.Microcontroller: Arduino Uno or Nano. Bridge:USB OTG Cable (Phone type to USB-A female). Sensor Hub: Any Android Phone (Android 8.0+). Compute: A Laptop/PC on the same Wi-Fi network.

💻 Software Stack1.

The Brain (PC)Language: Python 3.10+Core Libs: aiortc (WebRTC), opencv-python (Vision), filterpy (Kalman Filter), flask-socketio (Dashboard).

Responsibilities:Receives video & sensor packets.Runs Lane Detection & Visual Odometry.Fuses GPS + IMU + VO into a coherent pose $(x, y, \theta)$.Calculates steering/throttle commands.

2. The Relay (Android App)Language:

Kotlin (Jetpack Compose).Libs: stream-webrtc-android, usb-serial-for-android. Responsibilities:Captures Camera2 frames.Listens to Accelerometer/Gyroscope/Magnetometer & Location Services.Forwards serial commands from WebRTC $\rightarrow$ USB OTG $\rightarrow$ Arduino.

3. The Muscle (Arduino)Language: C++.

Responsibilities:Parses serial strings (e.g., CMD,150,-150).Generates PWM signals for motor drivers.Implements safety watchdog (stops if phone disconnects).

⚡ Getting Started

Step 1: Arduino SetupOpen Arduino/Aiduinocontrol.ino. Connect your Arduino. Upload the sketch. Wiring:Left Motor PWM: Pin 6 (En), Pins 4, 5 (Dir)Right Motor PWM: Pin 9 (En), Pins 7, 8 (Dir) Step 2: Android SetupOpen robotcontroller in Android Studio. Build and Install the CellBot Controller app on your phone. Permissions: Allow Camera, Location, and USB access when prompted. Step 3: PC Brain SetupCritical: You must run the Signaling Server and the Main Brain in two separate terminals. Install dependencies:Bashpip install -r brain/requirements.txt Terminal 1: The Switchboard (Keep this running!) signaling_server.py Wait until you see: Starting signaling server on 0.0.0.0:8765Terminal 2: The /main.py Step 4: Launch 🚀Connect your Phone to the Arduino via OTG cable. Open the App on the phone. Enter your PC's local IP address (e.g., 192.168.1.10) and click Connect. Open http://localhost:5000 on your PC to view the Dashboard. Click Manual to drive with arrow keys, or set a waypoint on the map for Autonomous mode. ##🛡️ Safety FeaturesWatchdog Timer If the PC or Phone crashes, the Arduino stops motors after 500ms of silence. Geofencing: Define safe zones on the dashboard map; robot stops if it breaches the boundary. Emergency Stop: Big red button on the Dashboard interface.

🤝 Contributing

Contributions are welcome! Please fork this repository and submit a pull request for improvements to the CV algorithms, EKF tuning, or Dashboard UI.

##📄 License This project is licensed under the MIT License - see the LICENSE file for details.