A self-driving car that uses neural networks and machine learning to move without colliding with other cars or objects.
Here is an example of the neural network passing with success all the cars on the road without colliding ๐
- Open the url: https://pncsoares.github.io/self-driving-car/src
- Watch the car driving like a Tesla but without colliding to other vehicles or walls ๐
- Click the save button when you see a good car* so the next generation inherit it's brain and mutate to the next generations
- Refresh the tab
- Redo this steps as many as you want ๐ช
good car : the one that avoids more traffic and reaches the farthest distance
๐ง The more times you run, the better the AI becomes!
- Car driving mechanics
- Road definition
- Artificial sensors
- Collision detection
- Traffic simulation
- Neural network
- Visualizing neural networks
- Parallelize AI
- Save best car's brain
- Genetic algorithm
- Mutation
Create and go to the directory where you want to place the repository
cd my-directory
Clone the project
git clone https://github.com/pncsoares/self-driving-car.git
Go to the project directory
cd self-driving-car
Open in Visual Studio Code
code .
MIT