Safe-Control-Gym: A Unified Benchmark Suite for Safe Learning-Based Control and Reinforcement Learning in Robotics

In recent years, both reinforcement learning and learning-based control-as well as the study of their safety , which is crucial for deployment in real-world robots-have gained significant traction. However, to adequately gauge the progress and applicability of new results, we need the tools to equit...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE robotics and automation letters Ročník 7; číslo 4; s. 11142 - 11149
Hlavní autoři: Yuan, Zhaocong, Hall, Adam W., Zhou, Siqi, Brunke, Lukas, Greeff, Melissa, Panerati, Jacopo, Schoellig, Angela P.
Médium: Journal Article
Jazyk:angličtina
Vydáno: Piscataway IEEE 01.10.2022
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Benchmark testing
Benchmarks
Control tasks
Learning
Machine learning for robot control
Open source software
Reinforcement learning
Reproducibility
Robot control
robot safety
Robotics
Robots
Safety
software tools for benchmarking and reproducibility
Task analysis
Trajectory control
ISSN:2377-3766, 2377-3766
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In recent years, both reinforcement learning and learning-based control-as well as the study of their safety , which is crucial for deployment in real-world robots-have gained significant traction. However, to adequately gauge the progress and applicability of new results, we need the tools to equitably compare the approaches proposed by the controls and reinforcement learning communities. Here, we propose a new open-source benchmark suite, called safe-control-gym , supporting both model-based and data-based control techniques. We provide implementations for three dynamic systems-the cart-pole, the 1D, and 2D quadrotor-and two control tasks-stabilization and trajectory tracking. We propose to extend OpenAI's Gym API-the de facto standard in reinforcement learning research-with (i) the ability to specify (and query) symbolic dynamics and (ii) constraints, and (iii) (repeatably) inject simulated disturbances in the control inputs, state measurements, and inertial properties. To demonstrate our proposal and in an attempt to bring research communities closer together, we show how to use safe-control-gym to quantitatively compare the control performance, data efficiency, and safety of multiple approaches from the fields of traditional control, learning-based control, and reinforcement learning.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2377-3766
2377-3766
DOI:10.1109/LRA.2022.3196132