Project: MADDPG for Collaboration

Different from single agent model, agents in this project should be capable to collaborate to max both scores as much as possible.

1. Unity Tennis

a 2D tennis environment.

1.1 Reward

In this environment, two agents control rackets to bounce a ball over a net. If an agent hits the ball over the net, it receives a reward of +0.1. If an agent lets a ball hit the ground or hits the ball out of bounds, it receives a reward of -0.01. Thus, the goal of each agent is to keep the ball in play. And this is a collaboration model.

1.2 State Space

The observation space consists of 8 variables corresponding to the position and velocity of the ball and racket. Each agent receives its own, local observation. Number of stacked vector observation is 3.

a single observation s=[x_ball, y_ball, vx_ball, vy_ball, x_racket, y_racket, vx_racket, vy_racket]. need to confirm the order and axis setup.

1.3 Action Space

Two continuous actions are available, corresponding to movement toward (or away from) the net, and jumping.

a single action a=[move_horizon, move_verticle]. all actions are between -1 and 1.

1.4 Task

The task is episodic, and in order to solve the environment, your agents must get an average score of +0.5 (over 100 consecutive episodes, after taking the maximum over both agents). Specifically,

After each episode, we add up the rewards that each agent received (without discounting), to get a score for each agent. This yields 2 (potentially different) scores. We then take the maximum of these 2 scores.
This yields a single score for each episode.

The environment is considered solved, when the average (over 100 episodes) of those scores is at least +0.5.

1.5 Environment

Download the environment from one of the links below. You need only select the environment that matches your operating system:
- Linux: click here
- Mac OSX: click here
- Windows (32-bit): click here
- Windows (64-bit): click here
Place the file in the root folder of GitHub repository, and unzip (or decompress) the file.
The environment was modified by Udacity, and it is not identical to original unity tennis environment.

1.6 Train & Evaluation

train script
```
python train_tennis.py
```
evaluate script
```
python eval_tennis.py
```

masszhou/unity_tennis