const env = new metacar.env("canvas", metacar.level.fullCity);
env.load();
Learn how to start with metacar by reading the documentation.
Algorithm: Q-learning with table
In this example, the goal is to demonstrate a simple working example of a reinforcement learning
algorithm using the metacar environment. You can check out
the demo or take a look at the medium tutorial.
Algorithm: Policy gradient algorithm based on Monte Carlo exploration (REINFORCE).
The neural network is build using tensorflow.js. Check out the demo.
The control is based on two continuous values for the throttle and steering angle of the car. The model is made using tensorflow.js
The control is based on two continuous values for the throttle and steering angle of the car.
The model is made using tensorflow.js
Metacar lets you edit your own level using the the editor. Also, you’re more than welcome to contribute to the project.