diff --git a/README.md b/README.md
index 3b1653c..66c13af 100644
--- a/README.md
+++ b/README.md
@@ -1,14 +1,32 @@
 # PG Travel
 PyTorch implementation of Vanilla Policy Gradient, Truncated Natural Policy Gradient, Trust Region Policy Optimization, Proximal Policy Optimization
 
+# Environment
+We have train PG agents in following environment
+* mujoco-py: [https://github.com/openai/mujoco-py](https://github.com/openai/mujoco-py)
+* Unity ml-agent walker: [https://github.com/Unity-Technologies/ml-agents](https://github.com/Unity-Technologies/ml-agents)
+
+# Requirements
+* python == 3.6
+* numpy
+* pytorch == 0.4
+* mujoco-py
+* ml-agent
+
 # Train
-* **algorithm**: PG, NPG, TRPO, PPO
+## 1. mujoco-py
+* **algorithm**: PG, TNPG, TRPO, PPO
 * **env**: Ant-v2, HalfCheetah-v2, Hopper-v2, Humanoid-v2, HumanoidStandup-v2, InvertedPendulum-v2, Reacher-v2, Swimmer-v2, Walker2d-v2
 ~~~
-python train.py --algorithm "algorithm name" --env "environment name"
+python mujoco/train.py --algorithm "algorithm name" --env "environment name"
 ~~~
 
 # Reference
 This code is modified version of codes
 * [OpenAI Baseline](https://github.com/openai/baselines/tree/master/baselines/trpo_mpi)
 * [Pytorch implemetation of TRPO](https://github.com/ikostrikov/pytorch-trpo)
+
+
+# Trained Agent
+* hopper
+![image](/img/hopper.gif)
\ No newline at end of file
diff --git a/img/hopper.gif b/img/hopper.gif
new file mode 100644
index 0000000..e33a319
Binary files /dev/null and b/img/hopper.gif differ