FPV RC Car with Autonomous Driving

A cutting-edge radio-controlled car platform combining first-person view driving experience with machine learning-powered autonomous navigation capabilities

Project Overview

FPV RC Car

The YugTroniX FPV RC Car with Autonomous Driving represents our innovative approach to combining traditional radio-controlled vehicles with modern autonomous navigation technology. This project integrates computer vision, machine learning, and robotics to create an educational platform for experimenting with self-driving vehicle concepts.

What makes our RC car unique is its dual-mode operation: users can either drive the car manually with a first-person perspective through a real-time video feed, or engage the autonomous mode where the vehicle navigates independently using its onboard sensors and AI algorithms. This versatility makes it an excellent tool for learning about autonomous driving systems, computer vision, and control theory.

Key Features

  • First-Person View System: Low-latency HD camera with wireless video transmission providing real-time driver perspective
  • Autonomous Navigation: Machine learning algorithms that enable the car to navigate environments without human input
  • Multi-sensor Integration: Fusion of camera data, ultrasonic sensors, and IMU for comprehensive environmental awareness
  • Lane Detection: Computer vision system that identifies and follows lane markings
  • Obstacle Avoidance: Real-time detection and avoidance of static and moving obstacles
  • Remote Monitoring: Web interface for viewing sensor data, video feed, and control parameters
  • Dual Control Modes: Seamless switching between manual FPV control and autonomous operation
  • Custom Chassis Design: 3D-printed components with optimized weight distribution and sensor mounting points

Technical Specifications

Hardware Components

Raspberry Pi 4 Arduino Nano HD FPV Camera 5.8GHz Video Transmitter HC-SR04 Ultrasonic Sensors MPU-6050 IMU Brushed DC Motors L298N Motor Controllers 3S LiPo Battery

Software Stack

Python OpenCV TensorFlow Lite Arduino IDE Flask Web Server WebRTC ROS (Robot Operating System)

Performance Metrics

Top Speed: 25 km/h Operation Time: 40 minutes Sensor Range: 4 meters Video Latency: <100ms Autonomous Decision Rate: 10 Hz

Development Timeline

Initial Design

Concept development, component selection, and architectural planning

Chassis Development

3D modeling, printing, and assembly of the custom chassis

Electronics Integration

Circuit design, component wiring, and power management implementation

FPV System Setup

Camera mounting, video transmission testing, and latency optimization

Software Development

Computer vision algorithms, control software, and web interface creation

Machine Learning Integration

Data collection, model training, and neural network implementation

Testing & Optimization

Real-world testing, parameter tuning, and performance enhancement

Project Team

This project brought together team members with expertise in robotics, software development, electronics, and machine learning to create our innovative autonomous FPV RC car platform.

Team Member

B.Sri Harsha Vardhan/h3>

Project Lead

Team Member

Bhavani Sankar

Team Member

Pavan

Team Member

J.Latha Sri

t

Gallery

Applications

Our FPV RC Car with Autonomous Driving capabilities serves multiple practical applications:

  • Educational Platform: Hands-on learning tool for students studying robotics, computer vision, and artificial intelligence
  • Algorithm Testing: Test bed for developing and refining autonomous navigation algorithms
  • Remote Inspection: Exploration of areas that may be difficult or dangerous for humans to access
  • Competitive Racing: Platform for autonomous racing competitions and challenges
  • Computer Vision Research: Real-world testing environment for computer vision algorithms and techniques

Technical Challenges

Throughout development, our team overcame several key technical challenges:

  • Low-latency Video Transmission: Optimizing the FPV system to minimize lag between camera capture and display
  • Real-time Processing: Achieving sufficient processing speed for obstacle detection and decision-making on resource-constrained hardware
  • Environmental Adaptation: Creating algorithms robust enough to handle varying lighting conditions and terrain types
  • Power Management: Balancing power consumption between propulsion, sensors, and computational resources
  • Sensor Fusion: Effectively combining data from multiple sensors for accurate environmental modeling

Future Development

We're continuing to enhance our autonomous RC car with planned improvements including:

  • Implementation of SLAM (Simultaneous Localization and Mapping) for improved navigation
  • Integration of more sophisticated machine learning models for advanced decision-making
  • Addition of GPS for outdoor navigation capabilities
  • Development of multi-vehicle communication for cooperative tasks
  • Creation of a more intuitive control interface with augmented reality elements
  • Enhanced telemetry and data logging for performance analysis

Want to join our FPV RC Car project team?

Become a YugTroniX Club member and contribute to the future of autonomous vehicle technology!

Contact Us