Grid World

A basic example where the AI is trained to reach to the destination and avoid the obstacle in a grid world.

This example shows how to create an environment with visual observation and action mask.

We copied the environment from Unity ml-agents, and modified it to train in editor. Here are the results:

After training

The agent is able to move the with target while avoiding the obstable

Visual Observation

The input observation of the agent is the image from a top down camera as above

Go to Sourcecode

This Grid World example is one of the examples in the the UnityTensorflowKeras repository. Go to the repository from the link below to install it according to the instructions.

The Grid World example is located under Assets/UnityTensorflow/Examples/GridWorld directory.

For more information about this example, see Here.

Exercises

NA