Skip to main content

A OpenAI Gym Env for Rover with Arm

Project description

A OpenAI Gym Env for Rover with Arm

Project Overview

https://docs.google.com/presentation/d/1NnCJ13qy9eBprIPJsYDFgFuNXPIBvIbRBtx6BVz_si8/edit#slide=id.p3

Task1 - Move the cart near the table and pick up the object in the tray

Reward = 1 (when the object is picked up)

Reward = 0 (else)

Code with Sample Actions

import rover_arm
import gym
env = gym.make('rover-arm-pick-v0', render_mode = 'rgb_array')

observation = env.reset()
done = False

while not done:
    action = env.action_space.sample()
    observation, reward, done, info = env.step(action)
    img = env.render()
    # print(img.shape)
    print(action, observation, reward)
    
print(reward, done, info)

You can try to explore the environment and action space by controlling the bot using Keyboard.

You will have to install the dev version in local, and give keyboard access to terminal or IDE where code is being executed. To install in dev version you could do


pip install 'rover-arm[dev]'

Keyboard Controls

Rover

Up, Down, Left, Right Arrows to steer the Rover.

Arm

A, D -> Move the end-effector in X-axis

W, S -> Move the end-effector in Y-axis

Q, E -> Move the end-effector in Z-axis

-, + -> Open / Close the fingers of the robot arm

Note: W, S are also hot keys to adjust view in pybullet env (so ignore the changes or press again to undo the change.)

Code to control the bot using keyboard in human mode (needs to be run in local)

import rover_arm.keyboard_control as kc

import rover_arm
import gym
env = gym.make('rover-arm-pick-v0', render_mode = 'human')

keyboard_controller = kc.KeyboardAction()
keyboard_controller.start_listening()

observation = env.reset()
done = False

while not done:
    action = keyboard_controller.action
    observation, reward, done, info = env.step(action)
    # print(action, observation, reward)
# print(reward, done, info)

Task2: Pick the object from the closer tray and place it on the distant tray

Use the env "rover-arm-place-v0" for task2


env = gym.make('rover-arm-place-v0')

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

rover_arm-1.0.10.tar.gz (2.7 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

rover_arm-1.0.10-py3-none-any.whl (2.8 MB view details)

Uploaded Python 3

File details

Details for the file rover_arm-1.0.10.tar.gz.

File metadata

  • Download URL: rover_arm-1.0.10.tar.gz
  • Upload date:
  • Size: 2.7 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.9.13

File hashes

Hashes for rover_arm-1.0.10.tar.gz
Algorithm Hash digest
SHA256 2ab57596d18d3b6365e10a1190e044771f81afa216c6008fdf87a27ad0c65d38
MD5 230930d19f4a00cf7049b8b0454f705b
BLAKE2b-256 f1451b20650e9c3b4eb91f9c7dac9ac75e27c175e32b3c2904fe3fd6ccc12d96

See more details on using hashes here.

File details

Details for the file rover_arm-1.0.10-py3-none-any.whl.

File metadata

  • Download URL: rover_arm-1.0.10-py3-none-any.whl
  • Upload date:
  • Size: 2.8 MB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.9.13

File hashes

Hashes for rover_arm-1.0.10-py3-none-any.whl
Algorithm Hash digest
SHA256 660ef7e9323f7853eb1f47478764fe15a0200ecd94c95b2dd32cc406e924d6d1
MD5 365746d3b635dc31fccb97dde737d63f
BLAKE2b-256 06b81e5e3f672729673c1623c508d6d39e23cd968412227791cd1b32c29ffc0f

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page