A Control Barrier Perspective on Episodic Learning via Projection-to-State Safety

@article{Taylor2020ACB,
  title={A Control Barrier Perspective on Episodic Learning via Projection-to-State Safety},
  author={A. J. Taylor and A. Singletary and Yisong Yue and A. Ames},
  journal={IEEE Control Systems Letters},
  year={2020},
  volume={5},
  pages={1019-1024}
}
  • A. J. Taylor, A. Singletary, +1 author A. Ames
  • Published 2020
  • Engineering, Computer Science
  • IEEE Control Systems Letters
  • In this letter we seek to quantify the ability of learning to improve safety guarantees endowed by Control Barrier Functions (CBFs). In particular, we investigate how model uncertainty in the time derivative of a CBF can be reduced via learning, and how this leads to stronger statements on the safe behavior of a system. To this end, we build upon the idea of Input-to-State Safety (ISSf) to define Projection-to-State Safety (PSSf), which characterizes degradation in safety in terms of a… CONTINUE READING

    Figures and Topics from this paper.

    Control Barrier Functions for Sampled-Data Systems with Input Delays
    1
    Learning Control Barrier Functions from Expert Demonstrations
    2

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 27 REFERENCES
    A Control Lyapunov Perspective on Episodic Learning via Projection to State Stability
    5
    End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Tasks
    55
    Robust Regression for Safe Exploration in Control
    10
    Learning for Safety-Critical Control with Control Barrier Functions
    12
    Barrier-Certified Adaptive Reinforcement Learning With Applications to Brushbot Navigation
    13
    Input-to-State Safety With Control Barrier Functions
    23
    A General Safety Framework for Learning-Based Control in Uncertain Robotic Systems
    94
    Episodic Learning with Control Lyapunov Functions for Uncertain Robotic Systems*
    19
    An Online Approach to Active Set Invariance
    13
    Safe Learning of Quadrotor Dynamics Using Barrier Certificates
    36