Goal:
Create an animation of a self-driving car that navigates randomly generated traffic. The car will have sensors that enable it to "see" traffic and road boundaries. Readings from the sensors are fed into a simple neural network that outputs movement controls for the car. Finally a genetic algorithm is used to search the solution space for a car with the right weights to navigate traffic confidently.
- Set up a repo
- Implement basic object classes - vehicle, car, road
- Implement vehicle physics
- Implement basic animation
- Implement sensors for cars
- Implement neural net "brain" on playerCar
- Implement neural network visualizer
- Parallelize cars
- Implement genetic algorithm for optimizing NN