Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/edluffy/tiago_gym
🦾 A ROS package to run Reinforcement Learning experiments, particularly pick and place tasks, on the TIAGo robot.
https://github.com/edluffy/tiago_gym
gazebo openai-gym reinforcement-learning robotics ros tiago-robot
Last synced: 24 days ago
JSON representation
🦾 A ROS package to run Reinforcement Learning experiments, particularly pick and place tasks, on the TIAGo robot.
- Host: GitHub
- URL: https://github.com/edluffy/tiago_gym
- Owner: edluffy
- License: mit
- Created: 2021-07-16T14:39:00.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2021-08-10T12:36:16.000Z (over 3 years ago)
- Last Synced: 2024-11-06T06:14:43.036Z (2 months ago)
- Topics: gazebo, openai-gym, reinforcement-learning, robotics, ros, tiago-robot
- Language: Python
- Homepage:
- Size: 557 KB
- Stars: 7
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 🦾 tiago_gym
A ROS package to run Reinforcement Learning experiments, particularly pick and place tasks, on the TIAGo robot. Uses Gazebo, Rviz and MoveIt! (for motion planning)
## Installation
Tested on Ubuntu 18.04 only. Beware! Instructions assume familiarity with the ROS [packages](http://wiki.ros.org/Packages) system.
- Install ROS Melodic + TIAGo
- http://wiki.ros.org/Robots/TIAGo/Tutorials/Installation/InstallUbuntuAndROS
- Install openai_ros package into your TIAGo workspace
``` bash
cd /home/user/tiago_public_ws/src
git clone https://bitbucket.org/theconstructcore/openai_ros.git
cd openai_ros;git checkout version2
cd /home/user/tiago_public_ws;catkin_make;source devel/setup.bash
```
- Install tiago_gym package into your TIAGo workspace
``` bash
cd /home/user/tiago_public_ws/src
git clone https://github.com/edluffy/tiago_gym
cd /home/user/tiago_public_ws;catkin_make;source devel/setup.bash
```
- Launch an environment!
- `roslaunch tiago_gym start_gym.launch`
- Gazebo and Rviz should launch similarly to gif above.
- To speed up simulation you can run the command `gz physics –u 0 -s 0.0025` in a separate terminal.
- Running custom training scripts
- Launch with `roslaunch tiago_gym start_gym.launch train:=false`
- Run in a separate terminal `python my_training_script.py`. Example snippets are shown in the [Environments](#environments)
section.
## Environments
| TiagoSimpleEnv-v0 | TiagoReachEnv-v0 |
| ------------------------ | ------------------------ |
 | |
### TiagoSimpleEnv-v0
This is a simple test environment in which the robot gripper must move to a discrete goal position in 3D space (essentially a 3D gridworld). Example usage:
``` python
import tiago_simple_env
from agents import dqn
env = gym.make('TiagoSimpleEnv-v0')
o_dims = len(env.observation_space.sample())
a_dims = env.action_space.n
agent = dqn.DQN(env, input_size=o_dims,
output_size=o_dims, alpha=0.01, epsilon_decay=0.95)
agent.run(100)
```
Observations ActionsRewards(Dense!)
| | |
| --- | ---------------------- |
| 0 | x-pos of gripper |
| 1 | y-pos of gripper |
| 2 | z-pos of gripper |
| | |
| --- | ---------------------- |
| 0 | x-pos of gripper + 0.1 |
| 1 | x-pos of gripper - 0.1 |
| 2 | y-pos of gripper + 0.1 |
| 3 | y-pos of gripper - 0.1 |
| 4 | z-pos of gripper + 0.1 |
| 5 | z-pos of gripper - 0.1 |
| | |
| -----------------|------------------ |
| Goal within 0.05 | 10 |
| Else | -Distance to goal |
### TiagoReachEnv-v0
A continuous action environment – robot can move a vector distance in any direction to get to the goal. Example usage:
``` python
import tiago_reach_env
from agents import ddpg
env = gym.make('TiagoReachEnv-v0')
o_dims = len(env.observation_space.sample())
a_dims = env.action_space.shape[0]
a_high = env.action_space.high
agent = ddpg.DDPG(env, o_dims, a_dims, a_high)
agent.run(100)
```
Observations ActionsRewards(Dense!)
| | |
| --- | --------------------------------------- |
| 0 | absolute pos of gripper |
| 1 | relative pos of gripper |
| | |
| --- | ---------------------- |
| 0 | x-pos of gripper |
| 1 | x-pos of gripper |
| 2 | y-pos of gripper |
| | |
| -----------------|------------------ |
| Goal within 0.05 | 10 |
| Else | -Distance to goal |
## Agents
- Tensorflow implementations of DQN and DDPG can be found in `scripts/agents`.