Code for "Adversarial Motion Priors Make Good Substitutes for Complex Reward Functions"

Overview

Adversarial Motion Priors Make Good Substitutes for Complex Reward Functions

Codebase for the "Adversarial Motion Priors Make Good Substitutes for Complex Reward Functions" project. This repository contains the code necessary to ground agent skills using small amounts of reference data (4.5 seconds). All experiments are performed using the A1 robot from Unitree. This repository is based off of Nikita Rudin's legged_gym repo, and enables us to train policies using Isaac Gym.

Maintainer: Alejandro Escontrela Affiliation: University of California at Berkeley Contact: [email protected]

Useful Links

Project website: https://bit.ly/3hpvbD6 Paper: https://drive.google.com/file/d/1kFm79nMmrc0ZIiH0XO8_HV-fj73agheO/view?usp=sharing

Installation

  1. Create a new python virtual env with python 3.6, 3.7 or 3.8 (3.8 recommended). i.e. with conda:
    • conda create -n amp_hw python==3.8
    • conda activate amp_hw
  2. Install pytorch 1.10 with cuda-11.3:
    • pip3 install torch==1.10.0+cu113 torchvision==0.11.1+cu113 tensorboard==2.8.0 pybullet==3.2.1 opencv-python==4.5.5.64 torchaudio==0.10.0+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html
  3. Install Isaac Gym
    • Download and install Isaac Gym Preview 3 (Preview 2 will not work!) from https://developer.nvidia.com/isaac-gym
    • cd isaacgym/python && pip install -e .
    • Try running an example cd examples && python 1080_balls_of_solitude.py
    • For troubleshooting check docs isaacgym/docs/index.html)
  4. Install rsl_rl (PPO implementation)
    • Clone this repository
    • cd AMP_for_hardware/rsl_rl && pip install -e .
  5. Install legged_gym
    • cd ../ && pip install -e .

CODE STRUCTURE

  1. Each environment is defined by an env file (legged_robot.py) and a config file (legged_robot_config.py). The config file contains two classes: one conatianing all the environment parameters (LeggedRobotCfg) and one for the training parameters (LeggedRobotCfgPPo).
  2. Both env and config classes use inheritance.
  3. Each non-zero reward scale specified in cfg will add a function with a corresponding name to the list of elements which will be summed to get the total reward. The AMP reward parameters are defined in LeggedRobotCfgPPO, as well as the path to the reference data.
  4. Tasks must be registered using task_registry.register(name, EnvClass, EnvConfig, TrainConfig). This is done in envs/__init__.py, but can also be done from outside of this repository.
  5. Reference data can be found in the datasets folder.

Usage

  1. Train:
    ```python legged_gym/scripts/train.py --task=a1_amp``
    • To run on CPU add following arguments: --sim_device=cpu, --rl_device=cpu (sim on CPU and rl on GPU is possible).
    • To run headless (no rendering) add --headless.
    • Important: To improve performance, once the training starts press v to stop the rendering. You can then enable it later to check the progress.
    • The trained policy is saved in AMP_for_hardware/logs/<experiment_name>/<date_time>_<run_name>/model_<iteration>.pt. Where <experiment_name> and <run_name> are defined in the train config.
    • The following command line arguments override the values set in the config files:
    • --task TASK: Task name.
    • --resume: Resume training from a checkpoint
    • --experiment_name EXPERIMENT_NAME: Name of the experiment to run or load.
    • --run_name RUN_NAME: Name of the run.
    • --load_run LOAD_RUN: Name of the run to load when resume=True. If -1: will load the last run.
    • --checkpoint CHECKPOINT: Saved model checkpoint number. If -1: will load the last checkpoint.
    • --num_envs NUM_ENVS: Number of environments to create.
    • --seed SEED: Random seed.
    • --max_iterations MAX_ITERATIONS: Maximum number of training iterations.
  2. Play a trained policy:
    python legged_gym/scripts/play.py --task=a1_amp
    • By default the loaded policy is the last model of the last run of the experiment folder.
    • Other runs/model iteration can be selected by setting load_run and checkpoint in the train config.
  3. Record video of a trained policy python legged_gym/scripts/record_policy.py --task=a1_amp
    • This saves a video of the in the base directory.

Adding a new environment

The base environment legged_robot implements a rough terrain locomotion task. The corresponding cfg does not specify a robot asset (URDF/ MJCF) and no reward scales.

  1. Add a new folder to envs/ with '<your_env>_config.py, which inherit from an existing environment cfgs
  2. If adding a new robot:
    • Add the corresponding assets to resourses/.
    • In cfg set the asset path, define body names, default_joint_positions and PD gains. Specify the desired train_cfg and the name of the environment (python class).
    • In train_cfg set experiment_name and run_name
  3. (If needed) implement your environment in <your_env>.py, inherit from an existing environment, overwrite the desired functions and/or add your reward functions.
  4. Register your env in isaacgym_anymal/envs/__init__.py.
  5. Modify/Tune other parameters in your cfg, cfg_train as needed. To remove a reward set its scale to zero. Do not modify parameters of other envs!

Troubleshooting

  1. If you get the following error: ImportError: libpython3.8m.so.1.0: cannot open shared object file: No such file or directory, do: sudo apt install libpython3.8

Known Issues

  1. The contact forces reported by net_contact_force_tensor are unreliable when simulating on GPU with a triangle mesh terrain. A workaround is to use force sensors, but the force are propagated through the sensors of consecutive bodies resulting in an undesireable behaviour. However, for a legged robot it is possible to add sensors to the feet/end effector only and get the expected results. When using the force sensors make sure to exclude gravity from trhe reported forces with sensor_options.enable_forward_dynamics_forces. Example:
    sensor_pose = gymapi.Transform()
    for name in feet_names:
        sensor_options = gymapi.ForceSensorProperties()
        sensor_options.enable_forward_dynamics_forces = False # for example gravity
        sensor_options.enable_constraint_solver_forces = True # for example contacts
        sensor_options.use_world_frame = True # report forces in world frame (easier to get vertical components)
        index = self.gym.find_asset_rigid_body_index(robot_asset, name)
        self.gym.create_asset_force_sensor(robot_asset, index, sensor_pose, sensor_options)
    (...)

    sensor_tensor = self.gym.acquire_force_sensor_tensor(self.sim)
    self.gym.refresh_force_sensor_tensor(self.sim)
    force_sensor_readings = gymtorch.wrap_tensor(sensor_tensor)
    self.sensor_forces = force_sensor_readings.view(self.num_envs, 4, 6)[..., :3]
    (...)

    self.gym.refresh_force_sensor_tensor(self.sim)
    contact = self.sensor_forces[:, :, 2] > 1.
Owner
Alejandro Escontrela
On the first epoch.
Alejandro Escontrela
Copy online media to your USB pen by night and watch it on your daily commute

commute-tube commute-tube is your friend on your daily commute. It will download videos of your interest to your USB pen by night so that you're able

Matthias Küch 19 Mar 23, 2022
Downloads state flags from wikipedia for states/regions from all countries

world-state-flags Downloads state flags from wikipedia for states/regions from all countries This data is NOT curated Uses https://github.com/dr5hn/co

João Ribeiro Bezerra 2 Dec 15, 2022
This Program helps you download songs from the Spotify track's link you give in.

Spotify-Downloader-GUI This Program helps you download songs from the Spotify track's link you give in. It uses yt-dlp to download songs from Youtube.

Harish 12 Jun 14, 2022
Bulk Downloader for Reddit

saveddit is a bulk media downloader for reddit pip3 install saveddit Setting up authorization Register an application with Reddit Write down your clie

Pranav 136 Jan 03, 2023
MMDL (Mega Music Downloader) - A tool to easily download music.

mmdl - Mega Music Downloader What is mmdl ❓ TLDR: MMDL is a cli app which allows you to quickly and efficiently download one or multiple songs from Yo

techboy-coder 30 Dec 13, 2022
This is Yt Downloader. Coded with Python (my first repository)

Get Started Download & install Python first before using this software. Download Python Installing Python and Pytube Library (IMPORTANT) Installing Py

Qi 2 Oct 25, 2021
Code for "Adversarial Motion Priors Make Good Substitutes for Complex Reward Functions"

Adversarial Motion Priors Make Good Substitutes for Complex Reward Functions Codebase for the "Adversarial Motion Priors Make Good Substitutes for Com

Alejandro Escontrela 54 Dec 13, 2022
Tool to get Canvas cover videos from Spotify tracks.

Spotify Canvas Downloader Tool to get Canvas cover videos from Spotify tracks. ✨ Try it out Building Clone the repository git clone https://github.com

Gabriel 35 Dec 28, 2022
Desktop utility to download images/videos/music/text from various websites, and more

Desktop utility to download images/videos/music/text from various websites, and more

Kurt Bestor 11.2k Jan 08, 2023
一个在新番更新后第一时间在dmhy等BT下载站自动下载的小工具.

Anime Track 一个在新番更新后第一时间自动下载的小工具. 可以根据自定义的关键字在dmhy等BT下载站在搜索结果更新时将磁力链发送至aria2实现自动下载. 基本功能包含: 将BT下载站的某个关键字的搜索结果的所有磁力链添加至ARIA2 自动更新aria2 trackers 将已添加的磁力

Sunky 24 Oct 12, 2022
A simple GUI video downloader built off of the python module 'yt-dlp'

Simple-Youtube-DL-Gui Supported Operating Systems Windows 7 (x64), Windows 8 (x64), and Windows 10 (x64) How to use Main Gui Extract program from arch

12 Dec 30, 2022
Download images where login is required using har python and js

이미지 다운로드(har, python, js 사용) 로그인이 필요한 사이트에서 DevTools로 이미지를 다운받는 방법은 조금 까다로웠다. 가장 쉽게 할 수 있는 방법을 찾아보았다. 사용법 F12를 눌러 DevTools를 실행 Network 탭으로 이동 페이지 새로고침

0 Jul 22, 2022
YoutubeDownloader - Download any public Playlist from Youtube

YoutubeDownloader Download any public Youtube Channel / Playlist Features Bulk d

17 Nov 12, 2022
YouPlay is a python based tool for downloading YouTube videos through its URL

YouPlay is a python based tool for downloading YouTube videos through its URL. It is capable to download videos from YouTube playlists too and can extract the audio file only from the video. It can r

Nitin Choudhury 10 Sep 15, 2022
Download Youtube videos in mp4 format in a fast, easy, convenient way made with Python!

yt_downloader Download Youtube videos in mp4 format in a fast, easy, convenient way made with Python! Required Modules pytube os time colorama Errors

3 Jul 02, 2022
Youtube_dl_helper - A hacky python script meant to automate the process of downloading mp3 files from YouTube using youtube-dl library

youtube_dl_helper A helper program meant to automate the process of downloading mp3 files from YouTube using youtube-dl library Dependencies In order

Guilherme Bittencourt de Borba 1 Jan 04, 2022
Super S3 command line tool

s4cmd Super S3 command line tool Author: Chou-han Yang (@chouhanyang) Current Maintainers: Debodirno Chandra (@debodirno) | Naveen Vardhi (@rozuur) |

Bloomreach 1.2k Dec 27, 2022
code for paper"3D reconstruction method based on a generative model in continuous latent space"

PyTorch implementation of 3D-VGT(3D-VAE-GAN-Transformer) This repository contains the source code for the paper "3D reconstruction method based on a g

Tong 5 Apr 25, 2022
the best video downloader for terminals (currently only compatible with Linux and Windows)

the best video downloader for terminals (currently only compatible with Linux and Windows)

Amaral 2 Oct 14, 2021
A downloader for the ISIS service of TU Berlin

isis_dl A downloading utility for the ISIS tool of TU-Berlin. Version 0.4 Features Downloads all Material from all courses of your ISIS page. Efficien

1 Nov 06, 2021