A low-cost autonomous robotic vehicle designed for surveillance, obstacle detection, and autonomous navigation - built from scratch without ROS, evolving towards full autonomy.
S.C.O.U.T. (Surveillance-Controlled Unmanned Tracker) is an open-source autonomous vehicle system that started as a remote-controlled robot and is evolving into a fully autonomous delivery prototype. The project emphasizes low-cost components, modular design, and step-by-step capability building.
- 🕹️ Dual Control Modes: Web-based UI and Manual RC control
- 📹 Real-time Video Streaming via ESP32-CAM
- 🚧 Obstacle Detection using ultrasonic sensors
- 📊 Sensor Monitoring Dashboard with real-time data
- 🧠 IMU Integration for orientation and acceleration tracking
- 🎮 Intuitive Control Interface with RC-style layout
- 🔄 Seamless Mode Switching between manual and web control
The project began with a clear vision: create a robust, tank-like robotic platform that could serve as the foundation for advanced autonomous capabilities. The name "SCOUT" perfectly captures its mission - a mobile system designed to scout environments and perform surveillance missions.
| Component | Purpose | Specifications |
|---|---|---|
| ESP32 | Main microcontroller | Web server, sensor processing, motor control |
| ESP32-CAM | Vision system | Real-time video streaming over Wi-Fi |
| HC-SR04 Ultrasonic Sensors | Obstacle detection | Front, left, and right positioning |
| MPU6050 | Inertial measurement | 6-axis gyroscope and accelerometer |
| L298N Motor Driver | Motor control | Dual H-bridge for differential drive |
| DC Motors | Locomotion | Geared motors for tank treads |
| FS-iA6B FlySky Receiver | RC control | 6-channel receiver for manual override |
| Tank Chassis | Platform | Robust base with tread system |
| Battery Pack | Power system | 12V Li-ion configuration |
1. Chassis Preparation
- Selected a tank-style chassis for maximum stability and terrain capability
- Mounted motor driver and ESP32 inside the chassis for protection
2. Sensor Integration
- Front Ultrasonic: Primary obstacle detection
- Side Ultrasonics: Lateral awareness for navigation
- MPU6050: Mounted for optimal orientation sensing
- Raspi-CAM: will be added
3. Control Electronics
- ESP32 configured as both web server and motor controller
- L298N motor driver wired for differential drive control
- FlySky receiver integrated for seamless manual override
- Power distribution system designed for stable operation
ESP32 (Main Controller)
├── Web Server (Control Interface)
├── Sensor Data Processing
├── Motor Control Logic
└── RC Signal Processing
Web-Based Control Interface
- Responsive 3x3 button grid layout
- Real-time sensor data display
Sensor Monitoring System
- Front/Back ultrasonic distance readings
- X/Y acceleration from MPU6050
- Calculated speed estimation
Dual Control Architecture
- Channel 6 Mode Switching: Seamless transition between RC and web control
- Manual Override: RC control takes precedence for safety
- Fail-safe Operation: Automatic stop on signal loss
- Access: Connect to ESP32's hosted Wi-Fi network
- Interface: Intuitive button layout (Forward, Back, Left, Right, Stop)
- Monitoring: Real-time sensor dashboard
- Transmitter: FlySky FS-i6 compatible
- Channels:
- CH1: Steering control
- CH2: Throttle control
- CH6: Mode switching (RC/Web)
- Override: Manual control always takes priority
- Obstacle Detection: will use camera
- Orientation Tracking: Real-time tilt and acceleration monitoring
- Speed Calculation: will use wheel encoders and IMU data
The current system uses simple reactive control based on sensor inputs:
- Obstacle avoidance using ultrasonic sensor thresholds
- Direct motor control based on user input or sensor triggers
# Planned PID controller for precise movement
class PIDController:
def __init__(self, kp, ki, kd):
self.kp = kp # Proportional gain
self.ki = ki # Integral gain
self.kd = kd # Derivative gain
# Implementation for wheel speed control and path following- State Variables: Position, velocity, orientation
- Control Inputs: Left/right motor speeds
- Optimization: Minimize tracking error and control effort
- Hardware Upgrade: Raspberry Pi 4 (8GB) as main computing unit
- Vision System: ToF camera or smartphone integration
- SLAM Implementation: Simultaneous Localization and Mapping
- Path Planning: A* or RRT algorithms for route optimization
- Object Recognition: Identify obstacles, targets, and landmarks
- Lane Detection: For structured environment navigation
- Visual Odometry: Camera-based position estimation
- Sensor Fusion: Combine visual and IMU data
- Point-to-Point Navigation: GPS and vision-based waypoint following
- Load Management: Payload handling and delivery confirmation
- Fleet Communication: Multi-robot coordination capabilities
This project is licensed under the MIT License - see the LICENSE file for details.
- GitHub Issues: For bug reports and feature requests
- Email: [pgnta.1385@gmail.com]
- Documentation: Comprehensive guides in the
/docsfolder
SCOUT represents more than just a robot - From basic remote control to advanced autonomous navigation, this project documents the complete journey of building a capable robotic system from scratch.
"Every expert was once a beginner. Every pro was once an amateur." - This project embodies that philosophy, showing that with persistence and proper documentation, anyone can build sophisticated robotic systems.