Train an agent using Stable Baselines

[duyminh1998]

We want to get an example RL agent working with our framework. One idea is to use Stable Baselines to kickstart something quick.

The current simple scenario is to train an RL agent to direct an aircraft to fly towards a set of coordinates.

[orgulous]

Split this into multiple tickets which show the steps to be done to demostrate multiple working examples of training agents using RL.