Provably Safe Reinforcement Learning via Action Projection Using Reachability Analysis and Polynomial Zonotopes

Provably Safe Reinforcement Learning via Action Projection Using Reachability Analysis and Polynomial Zonotopes

While reinforcement learning produces very promising results for many applications, its main disadvantage is the lack of safety guarantees, which prevents its use in safety-critical systems. In this work, we address this issue by a safety shield for nonlinear continuous systems that solve reach-avoi...

Full description

Bibliographic Details
Main Authors: Niklas Kochdumper, Hanna Krasowski, Xiao Wang, Stanley Bak, Matthias Althoff
Format: Article
Language:English
Published: IEEE 2023-01-01
Series:IEEE Open Journal of Control Systems
Subjects:
Action projection
reach-avoid problems
reachability analysis
reinforcement learning
Online Access:https://ieeexplore.ieee.org/document/10068193/

Internet

https://ieeexplore.ieee.org/document/10068193/

Similar Items