Automatically Composing and Parameterizing Skills by Evolving Finite State Automata

Lorenzo Riano, TM McGinnity

    Research output: Contribution to journalArticle

    14 Citations (Scopus)

    Abstract

    We propose a robotics algorithm that is able to simultaneously combine, adapt and create actions to solve a task. The actions are combined in a Finite State Automaton whose structure is determined by a novel evolutionary algorithm. The actions parameters, or new actions, are evolved alongside the FSA topology. Actions can be combined together in a hierarchical fashion. This approach relies on skills that with which the robot is already provided, like grasping or motion planning. Therefore software reuse is an important advantage of our proposed approach. We conducted several experiments both in simulation and on a real mobile manipulator PR2 robot, where skills of increasing complexity are evolved. Our results show that i) an FSA generated in simulation can be directly applied to a real robot without modifications and ii) the evolved FSA is robust to the noise and the uncertainty arising from real-world sensors.
    LanguageEnglish
    Pages639-650
    JournalRobotics and Autonomous Systems
    Volume60
    Issue number4
    DOIs
    Publication statusPublished - 10 Jan 2012

    Fingerprint

    Finite automata
    Robots
    Computer software reusability
    Motion planning
    Evolutionary algorithms
    Manipulators
    Robotics
    Topology
    Sensors
    Experiments

    Cite this

    @article{514b529eff554c88a91ba853912dc789,
    title = "Automatically Composing and Parameterizing Skills by Evolving Finite State Automata",
    abstract = "We propose a robotics algorithm that is able to simultaneously combine, adapt and create actions to solve a task. The actions are combined in a Finite State Automaton whose structure is determined by a novel evolutionary algorithm. The actions parameters, or new actions, are evolved alongside the FSA topology. Actions can be combined together in a hierarchical fashion. This approach relies on skills that with which the robot is already provided, like grasping or motion planning. Therefore software reuse is an important advantage of our proposed approach. We conducted several experiments both in simulation and on a real mobile manipulator PR2 robot, where skills of increasing complexity are evolved. Our results show that i) an FSA generated in simulation can be directly applied to a real robot without modifications and ii) the evolved FSA is robust to the noise and the uncertainty arising from real-world sensors.",
    author = "Lorenzo Riano and TM McGinnity",
    year = "2012",
    month = "1",
    day = "10",
    doi = "10.1016/j.robot.2012.01.002",
    language = "English",
    volume = "60",
    pages = "639--650",
    journal = "Robotics and Autonomous Systems",
    issn = "0921-8890",
    publisher = "Elsevier",
    number = "4",

    }

    Automatically Composing and Parameterizing Skills by Evolving Finite State Automata. / Riano, Lorenzo; McGinnity, TM.

    In: Robotics and Autonomous Systems, Vol. 60, No. 4, 10.01.2012, p. 639-650.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Automatically Composing and Parameterizing Skills by Evolving Finite State Automata

    AU - Riano, Lorenzo

    AU - McGinnity, TM

    PY - 2012/1/10

    Y1 - 2012/1/10

    N2 - We propose a robotics algorithm that is able to simultaneously combine, adapt and create actions to solve a task. The actions are combined in a Finite State Automaton whose structure is determined by a novel evolutionary algorithm. The actions parameters, or new actions, are evolved alongside the FSA topology. Actions can be combined together in a hierarchical fashion. This approach relies on skills that with which the robot is already provided, like grasping or motion planning. Therefore software reuse is an important advantage of our proposed approach. We conducted several experiments both in simulation and on a real mobile manipulator PR2 robot, where skills of increasing complexity are evolved. Our results show that i) an FSA generated in simulation can be directly applied to a real robot without modifications and ii) the evolved FSA is robust to the noise and the uncertainty arising from real-world sensors.

    AB - We propose a robotics algorithm that is able to simultaneously combine, adapt and create actions to solve a task. The actions are combined in a Finite State Automaton whose structure is determined by a novel evolutionary algorithm. The actions parameters, or new actions, are evolved alongside the FSA topology. Actions can be combined together in a hierarchical fashion. This approach relies on skills that with which the robot is already provided, like grasping or motion planning. Therefore software reuse is an important advantage of our proposed approach. We conducted several experiments both in simulation and on a real mobile manipulator PR2 robot, where skills of increasing complexity are evolved. Our results show that i) an FSA generated in simulation can be directly applied to a real robot without modifications and ii) the evolved FSA is robust to the noise and the uncertainty arising from real-world sensors.

    U2 - 10.1016/j.robot.2012.01.002

    DO - 10.1016/j.robot.2012.01.002

    M3 - Article

    VL - 60

    SP - 639

    EP - 650

    JO - Robotics and Autonomous Systems

    T2 - Robotics and Autonomous Systems

    JF - Robotics and Autonomous Systems

    SN - 0921-8890

    IS - 4

    ER -