A Robot to Shape your Natural Plant: The Machine Learning Approach to Model and Control Bio-Hybrid Systems

Mostafa Wahby, Mary Katherine Heinrich, Daniel Nicolas Hofstadler, Sebastian Risi, Payam Zahadat, Thomas Schmickl, Phil Ayres, Heiko Hamann

Publikation: Working paperForskning

4 Downloads (Pure)

Resumé

Bio-hybrid systems---close couplings of natural organisms with technology---are high potential and still underexplored. In existing work, robots have mostly influenced group behaviors of animals. We explore the possibilities of mixing robots with natural plants, merging useful attributes. Significant synergies arise by combining the plants' ability to efficiently produce shaped material and the robots' ability to extend sensing and decision-making behaviors. However, programming robots to control plant motion and shape requires good knowledge of complex plant behaviors. Therefore, we use machine learning to create a holistic plant model and evolve robot controllers. As a benchmark task we choose obstacle avoidance. We use computer vision to construct a model of plant stem stiffening and motion dynamics by training an LSTM network. The LSTM network acts as a forward model predicting change in the plant, driving the evolution of neural network robot controllers. The evolved controllers augment the plants' natural light-finding and tissue-stiffening behaviors to avoid obstacles and grow desired shapes. We successfully verify the robot controllers and bio-hybrid behavior in reality, with a physical setup and actual plants.
OriginalsprogEngelsk
Antal sider15
StatusUdgivet - 19 apr. 2018

Emneord

    Kunstnerisk udviklingsvirksomhed (KUV)

    • Nej

    Citer dette

    Wahby, M., Heinrich, M. K., Hofstadler, D. N., Risi, S., Zahadat, P., Schmickl, T., ... Hamann, H. (2018). A Robot to Shape your Natural Plant: The Machine Learning Approach to Model and Control Bio-Hybrid Systems.
    Wahby, Mostafa ; Heinrich, Mary Katherine ; Hofstadler, Daniel Nicolas ; Risi, Sebastian ; Zahadat, Payam ; Schmickl, Thomas ; Ayres, Phil ; Hamann, Heiko. / A Robot to Shape your Natural Plant : The Machine Learning Approach to Model and Control Bio-Hybrid Systems. 2018.
    @techreport{eceb4dd51e91420c8147e4174552364f,
    title = "A Robot to Shape your Natural Plant: The Machine Learning Approach to Model and Control Bio-Hybrid Systems",
    abstract = "Bio-hybrid systems---close couplings of natural organisms with technology---are high potential and still underexplored. In existing work, robots have mostly influenced group behaviors of animals. We explore the possibilities of mixing robots with natural plants, merging useful attributes. Significant synergies arise by combining the plants' ability to efficiently produce shaped material and the robots' ability to extend sensing and decision-making behaviors. However, programming robots to control plant motion and shape requires good knowledge of complex plant behaviors. Therefore, we use machine learning to create a holistic plant model and evolve robot controllers. As a benchmark task we choose obstacle avoidance. We use computer vision to construct a model of plant stem stiffening and motion dynamics by training an LSTM network. The LSTM network acts as a forward model predicting change in the plant, driving the evolution of neural network robot controllers. The evolved controllers augment the plants' natural light-finding and tissue-stiffening behaviors to avoid obstacles and grow desired shapes. We successfully verify the robot controllers and bio-hybrid behavior in reality, with a physical setup and actual plants.",
    keywords = "Bio-hybrid Systems, Machine Learning, LSTM Network, Plant Model, Evolved Robotic Controllers",
    author = "Mostafa Wahby and Heinrich, {Mary Katherine} and Hofstadler, {Daniel Nicolas} and Sebastian Risi and Payam Zahadat and Thomas Schmickl and Phil Ayres and Heiko Hamann",
    year = "2018",
    month = "4",
    day = "19",
    language = "English",
    type = "WorkingPaper",

    }

    A Robot to Shape your Natural Plant : The Machine Learning Approach to Model and Control Bio-Hybrid Systems. / Wahby, Mostafa; Heinrich, Mary Katherine; Hofstadler, Daniel Nicolas; Risi, Sebastian; Zahadat, Payam; Schmickl, Thomas; Ayres, Phil; Hamann, Heiko.

    2018.

    Publikation: Working paperForskning

    TY - UNPB

    T1 - A Robot to Shape your Natural Plant

    T2 - The Machine Learning Approach to Model and Control Bio-Hybrid Systems

    AU - Wahby, Mostafa

    AU - Heinrich, Mary Katherine

    AU - Hofstadler, Daniel Nicolas

    AU - Risi, Sebastian

    AU - Zahadat, Payam

    AU - Schmickl, Thomas

    AU - Ayres, Phil

    AU - Hamann, Heiko

    PY - 2018/4/19

    Y1 - 2018/4/19

    N2 - Bio-hybrid systems---close couplings of natural organisms with technology---are high potential and still underexplored. In existing work, robots have mostly influenced group behaviors of animals. We explore the possibilities of mixing robots with natural plants, merging useful attributes. Significant synergies arise by combining the plants' ability to efficiently produce shaped material and the robots' ability to extend sensing and decision-making behaviors. However, programming robots to control plant motion and shape requires good knowledge of complex plant behaviors. Therefore, we use machine learning to create a holistic plant model and evolve robot controllers. As a benchmark task we choose obstacle avoidance. We use computer vision to construct a model of plant stem stiffening and motion dynamics by training an LSTM network. The LSTM network acts as a forward model predicting change in the plant, driving the evolution of neural network robot controllers. The evolved controllers augment the plants' natural light-finding and tissue-stiffening behaviors to avoid obstacles and grow desired shapes. We successfully verify the robot controllers and bio-hybrid behavior in reality, with a physical setup and actual plants.

    AB - Bio-hybrid systems---close couplings of natural organisms with technology---are high potential and still underexplored. In existing work, robots have mostly influenced group behaviors of animals. We explore the possibilities of mixing robots with natural plants, merging useful attributes. Significant synergies arise by combining the plants' ability to efficiently produce shaped material and the robots' ability to extend sensing and decision-making behaviors. However, programming robots to control plant motion and shape requires good knowledge of complex plant behaviors. Therefore, we use machine learning to create a holistic plant model and evolve robot controllers. As a benchmark task we choose obstacle avoidance. We use computer vision to construct a model of plant stem stiffening and motion dynamics by training an LSTM network. The LSTM network acts as a forward model predicting change in the plant, driving the evolution of neural network robot controllers. The evolved controllers augment the plants' natural light-finding and tissue-stiffening behaviors to avoid obstacles and grow desired shapes. We successfully verify the robot controllers and bio-hybrid behavior in reality, with a physical setup and actual plants.

    KW - Bio-hybrid Systems

    KW - Machine Learning

    KW - LSTM Network

    KW - Plant Model

    KW - Evolved Robotic Controllers

    UR - https://arxiv.org/pdf/1804.06682.pdf

    M3 - Working paper

    BT - A Robot to Shape your Natural Plant

    ER -