Systems and methods are provided for navigating an autonomous vehicle using reinforcement learning techniques. In one implementation, a navigation system for a host vehicle may include at least one processing device programmed to: receive, from a camera, a plurality of images representative of an environment of the host vehicle; analyze the plurality of images to identify a present navigational state associated with the host vehicle; determine a first potential navigational action for the host vehicle based on the identified present navigational state; determine a first indicator of an expected reward based on the first potential navigational action and the identified present navigational state; predict a first future navigational state based on the first potential navigational action; determine a second indicator of an expected reward associated with at least one future action determined to be available to the host vehicle in response to the first future navigational state; determine a second potential navigational action for the host vehicle based on the identified present navigational state; determine a third indicator of an expected reward based on the second potential navigational action and the identified present navigational state; predict a second future navigational state based on the second potential navigational action; determine a fourth indicator of an expected reward associated with at least one future action determined to be available to the host vehicle in response to the second future navigational state; select the second potential navigational action based on a determination that, while the expected reward associated with the first indicator is greater than the expected reward associated with the third indicator, the expected reward associated with the fourth indicator is greater than the expected reward associated with the first indicator; and cause at least one adjustment of a navigational actuator of the host vehicle in response to the selected second potential navigational action.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    SUBOPTIMAL IMMEDIATE NAVIGATIONAL RESPONSE BASED ON LONG TERM PLANNING


    Beteiligte:
    STEIN GIDEON (Autor:in) / SHALEV-SHWARTZ SHAI (Autor:in) / SHAMMAH SHAKED (Autor:in) / SHASHUA AMNON (Autor:in)

    Erscheinungsdatum :

    2019-09-19


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Englisch


    Klassifikation :

    IPC:    G05D SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES , Systeme zum Steuern oder Regeln nichtelektrischer veränderlicher Größen / B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion / G01C Messen von Entfernungen, Höhen, Neigungen oder Richtungen , MEASURING DISTANCES, LEVELS OR BEARINGS / G06K Erkennen von Daten , RECOGNITION OF DATA / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Suboptimal immediate navigational response based on long term planning

    STEIN GIDEON / SHALEV-SHWARTZ SHAI / SHAMMAH SHAKED et al. | Europäisches Patentamt | 2020

    Freier Zugriff

    SUBOPTIMAL IMMEDIATE NAVIGATIONAL RESPONSE BASED ON LONG TERM PLANNING

    STEIN GIDEON / SHALEV-SHWARTZ SHAI / SHAMMAH SHAKED et al. | Europäisches Patentamt | 2018

    Freier Zugriff

    Suboptimal immediate navigational response based on long term planning

    STEIN GIDEON / SHALEV-SHWARTZ SHAI / SHAMMAH SHAKED et al. | Europäisches Patentamt | 2020

    Freier Zugriff

    Navigational Planning in Orienteering

    Murakoshi, Shin | Online Contents | 1997