Self-Driving Car

Overview

This project demonstrates a basic self-driving car simulation built from scratch using JavaScript without any external libraries. The simulation includes car driving mechanics, artificial sensors for environment detection, collision detection, a custom neural network, and a genetic algorithm for evolving the car's driving abilities.

Project Demo

Click the image to watch the video on YouTube.

Key Features

Car Driving Mechanics

• Implemented basic physics for acceleration, braking, and turning.
• Simple controls for manual testing and development purposes.

Artificial Sensors

• Each car is equipped with a set of sensors that emit rays in various directions.
• Sensors measure the distance to the nearest obstacle or car, providing crucial data for decision-making.

Collision Detection

• Developed a collision detection system that checks for overlaps between the car's bounding box and obstacles or other cars.
• Handles collision response to prevent the car from moving through obstacles.

Neural Network

• Custom-built neural network to process sensor inputs.
• Decides on the car's actions (accelerate, brake, turn) based on sensor data.

Genetic Algorithm

• Used a genetic algorithm to evolve the neural network.
• Selects the best-performing cars from each generation, mutates their neural networks, and creates new generations with improved performance.

Getting Started

Prerequisites

• A modern web browser (Chrome, Firefox, Safari)

Installation

1. Clone the repository:

   git clone https://github.com/danfirsten/SelfDriving-Car.git
   cd selfdriving-car

2. No external libraries are required.

Running the Simulation

1. Open index.html in your web browser to start the simulation.