Temporal logics are formal, expressive languages traditionally used in the computer science area of formal methods to specify the correctness of digital circuits and computer programs. Safety, as defined in previous chapters, is just a particular case of a temporal logic formula. In this chapter, we discuss adding general, temporal logic specifications to the control problem considered previously. In Sect. 9.1, we introduce Linear Temporal Logic (LTL) and automata accepting languages satisfying LTL formulas. Our approach to constructing controllers that enforce the satisfaction of an LTL formula is to break down an LTL formula into a sequence of reach-avoid subproblems. In Sect. 9.2, we formulate and solve these subproblems, and in Sect. 9.3 we present a hybrid system approach to combine the above controllers in such a way that the trajectories of the closed loop system satisfy the formula. We extend this procedure to systems for which CLFs are not known or for which the dynamics are uncertain in Sect. 9.4, where we leverage the MBRL framework from Chap. 8. We illustrate the method developed in this chapter with numerical examples in Sect. 9.5 and conclude with final remarks, references, and suggestions for further reading in Sect. 9.6.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Temporal Logic Guided Safe Model-Based Reinforcement Learning


    Weitere Titelangaben:

    synth. Lectures on Computer sci.


    Beteiligte:
    Cohen, Max (Autor:in) / Belta, Calin (Autor:in)


    Erscheinungsdatum :

    2023-05-16


    Format / Umfang :

    28 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Safe Exploration in Model-Based Reinforcement Learning

    Cohen, Max / Belta, Calin | Springer Verlag | 2023


    Safe robot execution in model-based reinforcement learning

    Martínez Martínez, David / Alenyà Ribas, Guillem / Torras, Carme | BASE | 2015

    Freier Zugriff

    Safe Trajectory Sampling in Model-Based Reinforcement Learning

    Zwane, S / Hadjivelichkov, D / Luo, Y et al. | BASE | 2023

    Freier Zugriff

    Safe robot execution in model-based reinforcement learning

    Martínez, David / Alenyà, Guillem / Torras, Carme | BASE | 2015

    Freier Zugriff

    Routing using Safe Reinforcement Learning

    Nayak Seetanadi, Gautham / Årzén, Karl-Erik | BASE | 2020

    Freier Zugriff