This project is an attempt to solve the reinforcement learning test environment called Reacher, which simulates 20 robotic arms in 3D and tasks the agent with controlling their movements in order to reach a specified target region. The required score of 30 (average return over 100 consecutive episodes) was achieved after 173 episodes, using a version of proximal policy optimisation.
In this environment, a double-jointed arm can move to target locations. A reward of +0.1 is provided for each step that the agent's hand is in the goal location. Thus, the goal of your agent is to maintain its position at the target location for as many time steps as possible.
The observation space consists of 33 variables corresponding to position, rotation, velocity, and angular velocities of the arm. Each action is a vector with four numbers, corresponding to torque applicable to two joints. Every entry in the action vector should be a number between -1 and 1.
This is the second version of the environment, which contains 20 identical agents, each with its own copy of the environment.
The agent must get an average score of +30 (over 100 consecutive episodes, and over all 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 20 (potentially different) scores. We then take the average of these 20 scores. This yields an average score for each episode (where the average is over all 20 agents).
If you haven't already, please follow the instructions in the DRLND GitHub repository to set up your Python environment. These instructions can be found in README.md at the root of the repository. By following these instructions, you will install PyTorch, the ML-Agents toolkit, and a few more Python packages required to complete the project.
(For Windows users) The ML-Agents toolkit supports Windows 10. While it might be possible to run the ML-Agents toolkit using other versions of Windows, it has not been tested on other versions. Furthermore, the ML-Agents toolkit has not been tested on a Windows VM such as Bootcamp or Parallels.
For this project, you will not need to install Unity - this is because we have already built the environment for you, and you can download it 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
Linux: click here Mac OSX: click here Windows (32-bit): click here Windows (64-bit): click here
Then, place the file in the p2_continuous-control/ folder in the DRLND GitHub repository, and unzip (or decompress) the file.
(For Windows users) Check out this link if you need help with determining if your computer is running a 32-bit version or 64-bit version of the Windows operating system.
(For AWS) If you'd like to train the agent on AWS (and have not enabled a virtual screen), then please use this link (version 1) or this link (version 2) to obtain the "headless" version of the environment. You will not be able to watch the agent without enabling a virtual screen, but you will be able to train the agent. (To watch the agent, you should follow the instructions to enable a virtual screen, and then download the environment for the Linux operating system above.)
Dependencies are listed in environment.yml. HoloViews and streamz are used to show live training progress, though these could be commented out if required.
To install:
conda env create --file environment.yml
To start the notebook:
jupyter notebook Continuous_Control_PPO.ipynb
Follow instructions in the notebook to train and run the agent.