Self-Driving Car Simulation

This project simulates a self-driving car using pure JavaScript. The simulation includes realistic car driving mechanics, a dynamic environment with obstacles, sensor data for decision-making, and a neural network controlling the car autonomously.

Features

• Realistic Car Driving Mechanics: Designed to enable smooth and realistic car movement.
• Dynamic Environment: Includes obstacles and boundaries for the car to navigate.
• Sensor Simulation: Detects obstacles and environmental features, providing input data for decision-making.
• Collision Detection: Efficient algorithms to ensure accurate responses to surroundings.
• Neural Network Control: Built from scratch to control the car autonomously using a genetic algorithm.
• Traffic Generation: Randomized traffic on the road to train the neural network based on interaction.
• Visualization Tools: Observe the neural network’s decision-making process in real-time.

Project Structure

• index.html: Main HTML file to set up the canvas and include scripts.
• style.css: Styles for the canvas and buttons.
• car.js: Defines the Car class and its behavior.
• controls.js: Handles user input for car controls.
• network.js: Defines the neural network and its operations.
• sensor.js: Defines the sensor class for obstacle detection.
• road.js: Defines the road and its properties.
• utils.js: Utility functions for linear interpolation and intersection detection.
• visualizer.js: Tools for visualizing the neural network.
• main.js: Main script to set up the simulation and handle animation.

How It Works

1. Car Mechanics: The car's movement is controlled by the Car class, which handles acceleration, friction, and turning based on user input or neural network output.
2. Environment: The Road class sets up the road with lanes and borders. The car navigates this environment, avoiding obstacles and staying within boundaries.
3. Sensors: The Sensor class simulates sensor rays that detect obstacles and provide input to the neural network.
4. Neural Network: The NeuralNetwork class controls the car based on sensor input. The network is trained through interaction with the environment and traffic.
5. Visualization: The Visualizer class provides real-time visualization of the neural network’s decision-making process.
6. Traffic Generation: Traffic is randomly generated in Main.js and used to train the car neural network.

Key Functions and Classes

• Car: Manages car properties and movement.
• Controls: Handles keyboard input for manual control.
• Road: Sets up the road and its properties.
• Sensor: Simulates sensors for obstacle detection.
• NeuralNetwork: Manages the neural network controlling the car.
• Visualizer: Provides tools for visualizing the neural network.

Screenshots