Awesome

safe-control-gym

Physics-based CartPole and Quadrotor Gym environments (using PyBullet) with symbolic a priori dynamics (using CasADi) for learning-based control, and model-free and model-based reinforcement learning (RL).

These environments include (and evaluate) symbolic safety constraints and implement input, parameter, and dynamics disturbances to test the robustness and generalizability of control approaches. [PDF]

@article{brunke2021safe,
         title={Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning},
         author={Lukas Brunke and Melissa Greeff and Adam W. Hall and Zhaocong Yuan and Siqi Zhou and Jacopo Panerati and Angela P. Schoellig},
         journal = {Annual Review of Control, Robotics, and Autonomous Systems},
         year={2021},
         url = {https://arxiv.org/abs/2108.06266}}

To reproduce the results in the article, see branch ar.

@article{yuan2021safecontrolgym,
  author={Yuan, Zhaocong and Hall, Adam W. and Zhou, Siqi and Brunke, Lukas and Greeff, Melissa and Panerati, Jacopo and Schoellig, Angela P.},
  journal={IEEE Robotics and Automation Letters},
  title={Safe-Control-Gym: A Unified Benchmark Suite for Safe Learning-Based Control and Reinforcement Learning in Robotics},
  year={2022},
  volume={7},
  number={4},
  pages={11142-11149},
  doi={10.1109/LRA.2022.3196132}}

To reproduce the results in the article, see branch submission.

Install on Ubuntu/macOS

Clone repo

git clone https://github.com/utiasDSL/safe-control-gym.git
cd safe-control-gym

(optional) Create a conda environment

Create and access a Python 3.10 environment using conda

conda create -n safe python=3.10
conda activate safe

Install

Install the safe-control-gym repository

python -m pip install --upgrade pip
python -m pip install -e .

Note

You may need to separately install gmp, a dependency of pycddlib:

conda install -c anaconda gmp

or

sudo apt-get install libgmp-dev

(optional) Additional requirements for MPC

You may need to separately install acados for fast MPC implementations.

• To build and install acados, see their installation guide.
• To set up the acados python interface, check out these installtion steps.

Architecture

Overview of safe-control-gym's API:

Configuration

Getting Started

Familiarize with APIs and environments with the scripts in examples/

3D Quadrotor Lemniscate Trajectory Tracking with PID

cd ./examples/   # Navigate to the examples folder
python3 pid/pid_experiment.py \
    --algo pid \
    --task quadrotor \
    --overrides \
        ./pid/config_overrides/quadrotor_3D/quadrotor_3D_tracking.yaml

<img src="figures/systems.png" alt="systems" width="450"> <img src="figures/figure8.gif" alt="trajectory" width="350">

Cartpole Stabilization with LQR

cd ./examples/   # Navigate to the examples folder
python3 lqr/lqr_experiment.py \
    --algo lqr \
    --task cartpole \
    --overrides \
        ./lqr/config_overrides/cartpole/cartpole_stabilization.yaml \
        ./lqr/config_overrides/cartpole/lqr_cartpole_stabilization.yaml

2D Quadrotor Trajectory Tracking with PPO

cd ./examples/rl/   # Navigate to the RL examples folder
python3 rl_experiment.py \
    --algo ppo \
    --task quadrotor \
    --overrides \
        ./config_overrides/quadrotor_2D/quadrotor_2D_track.yaml \
        ./config_overrides/quadrotor_2D/ppo_quadrotor_2D.yaml \
    --kv_overrides \
        algo_config.training=False

Verbose API Example

cd ./examples/   # Navigate to the examples folder
python3 no_controller/verbose_api.py \
    --task cartpole \
    --overrides no_controller/verbose_api.yaml

List of Implemented Controllers

• PID
• LQR
• iLQR
• Linear MPC
• GP-MPC
• SAC
• PPO
• DDPG
• Safety Layer
• RARL
• RAP

List of Implemented Safety Filters

• MPSC
• CBF
• Neural Network CBF

Performance

We compare the sample efficiency of safe-control-gym with the original OpenAI Cartpole and PyBullet Gym's Inverted Pendulum, as well as gym-pybullet-drones. We choose the default physic simulation integration step of each project. We report performance results for open-loop, random action inputs. Note that the Bullet engine frequency reported for safe-control-gym is typically much finer grained for improved fidelity. safe-control-gym quadrotor environment is not as light-weight as gym-pybullet-drones but provides the same order of magnitude speed-up and several more safety features/symbolic models.

^ Whether the environment includes a default set of constraints and disturbances

^^ Speed-up = Elapsed Simulation Time / Elapsed Wall Clock Time; on a 2.30GHz Quad-Core i7-1068NG7 with 32GB 3733MHz LPDDR4X; no GPU

Run Tests and Linting

Tests can be run locally by executing:

python3 -m pytest ./tests/  # Run all tests

Linting can be run locally with:

pre-commit install  # Install the pre-commit hooks
pre-commit autoupdate  # Auto-update the version of the hooks
pre-commit run --all  # Run the hooks on all files

References

• Brunke, L., Greeff, M., Hall, A. W., Yuan, Z., Zhou, S., Panerati, J., & Schoellig, A. P. (2022). Safe learning in robotics: From learning-based control to safe reinforcement learning. Annual Review of Control, Robotics, and Autonomous Systems, 5, 411-444.
• Yuan, Z., Hall, A. W., Zhou, S., Brunke, L., Greeff, M., Panerati, J., & Schoellig, A. P. (2022). safe-control-gym: A unified benchmark suite for safe learning-based control and reinforcement learning in robotics. IEEE Robotics and Automation Letters, 7(4), 11142-11149.

Related Open-source Projects

• gym-pybullet-drones: single and multi-quadrotor environments
• stable-baselines3: PyTorch reinforcement learning algorithms
• bullet3: multi-physics simulation engine
• gym: OpenAI reinforcement learning toolkit
• casadi: symbolic framework for numeric optimization
• safety-gym: environments for safe exploration in RL
• realworldrl_suite: real-world RL challenge framework
• gym-marl-reconnaissance: multi-agent heterogeneous (UAV/UGV) environments

University of Toronto's Dynamic Systems Lab / Vector Institute for Artificial Intelligence