Material recognition CNNs and hierarchical planning for biped robot locomotion on slippery terrain

Martim Brandão, Yukitoshi Minami Shiguematsu, Kenji Hashimoto, Atsuo Takanishi

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    13 Citations (Scopus)

    Abstract

    In this paper we tackle the problem of visually predicting surface friction for environments with diverse surfaces, and integrating this knowledge into biped robot locomotion planning. The problem is essential for autonomous robot locomotion since diverse surfaces with varying friction abound in the real world, from wood to ceramic tiles, grass or ice, which may cause difficulties or huge energy costs for robot locomotion if not considered. We propose to estimate friction and its uncertainty from visual estimation of material classes using convolutional neural networks, together with probability distribution functions of friction associated with each material. We then robustly integrate the friction predictions into a hierarchical (footstep and full-body) planning method using chance constraints, and optimize the same trajectory costs at both levels of the planning method for consistency. Our solution achieves fully autonomous perception and locomotion on slippery terrain, which considers not only friction and its uncertainty, but also collision, stability and trajectory cost. We show promising friction prediction results in real pictures of outdoor scenarios, and planning experiments on a real robot facing surfaces with different friction.

    Original languageEnglish
    Title of host publicationHumanoids 2016 - IEEE-RAS International Conference on Humanoid Robots
    PublisherIEEE Computer Society
    Pages81-88
    Number of pages8
    ISBN (Electronic)9781509047185
    DOIs
    Publication statusPublished - 2016 Dec 30
    Event16th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2016 - Cancun, Mexico
    Duration: 2016 Nov 152016 Nov 17

    Other

    Other16th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2016
    CountryMexico
    CityCancun
    Period16/11/1516/11/17

    Fingerprint

    Robots
    Friction
    Planning
    Trajectories
    Costs
    Tile
    Probability distributions
    Distribution functions
    Ice
    Wood
    Neural networks
    Experiments
    Uncertainty

    ASJC Scopus subject areas

    • Artificial Intelligence
    • Computer Vision and Pattern Recognition
    • Hardware and Architecture
    • Human-Computer Interaction
    • Electrical and Electronic Engineering

    Cite this

    Brandão, M., Shiguematsu, Y. M., Hashimoto, K., & Takanishi, A. (2016). Material recognition CNNs and hierarchical planning for biped robot locomotion on slippery terrain. In Humanoids 2016 - IEEE-RAS International Conference on Humanoid Robots (pp. 81-88). [7803258] IEEE Computer Society. https://doi.org/10.1109/HUMANOIDS.2016.7803258

    Material recognition CNNs and hierarchical planning for biped robot locomotion on slippery terrain. / Brandão, Martim; Shiguematsu, Yukitoshi Minami; Hashimoto, Kenji; Takanishi, Atsuo.

    Humanoids 2016 - IEEE-RAS International Conference on Humanoid Robots. IEEE Computer Society, 2016. p. 81-88 7803258.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Brandão, M, Shiguematsu, YM, Hashimoto, K & Takanishi, A 2016, Material recognition CNNs and hierarchical planning for biped robot locomotion on slippery terrain. in Humanoids 2016 - IEEE-RAS International Conference on Humanoid Robots., 7803258, IEEE Computer Society, pp. 81-88, 16th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2016, Cancun, Mexico, 16/11/15. https://doi.org/10.1109/HUMANOIDS.2016.7803258
    Brandão M, Shiguematsu YM, Hashimoto K, Takanishi A. Material recognition CNNs and hierarchical planning for biped robot locomotion on slippery terrain. In Humanoids 2016 - IEEE-RAS International Conference on Humanoid Robots. IEEE Computer Society. 2016. p. 81-88. 7803258 https://doi.org/10.1109/HUMANOIDS.2016.7803258
    Brandão, Martim ; Shiguematsu, Yukitoshi Minami ; Hashimoto, Kenji ; Takanishi, Atsuo. / Material recognition CNNs and hierarchical planning for biped robot locomotion on slippery terrain. Humanoids 2016 - IEEE-RAS International Conference on Humanoid Robots. IEEE Computer Society, 2016. pp. 81-88
    @inproceedings{03e663b89a0a41cea1e8ce96f4aa6053,
    title = "Material recognition CNNs and hierarchical planning for biped robot locomotion on slippery terrain",
    abstract = "In this paper we tackle the problem of visually predicting surface friction for environments with diverse surfaces, and integrating this knowledge into biped robot locomotion planning. The problem is essential for autonomous robot locomotion since diverse surfaces with varying friction abound in the real world, from wood to ceramic tiles, grass or ice, which may cause difficulties or huge energy costs for robot locomotion if not considered. We propose to estimate friction and its uncertainty from visual estimation of material classes using convolutional neural networks, together with probability distribution functions of friction associated with each material. We then robustly integrate the friction predictions into a hierarchical (footstep and full-body) planning method using chance constraints, and optimize the same trajectory costs at both levels of the planning method for consistency. Our solution achieves fully autonomous perception and locomotion on slippery terrain, which considers not only friction and its uncertainty, but also collision, stability and trajectory cost. We show promising friction prediction results in real pictures of outdoor scenarios, and planning experiments on a real robot facing surfaces with different friction.",
    author = "Martim Brand{\~a}o and Shiguematsu, {Yukitoshi Minami} and Kenji Hashimoto and Atsuo Takanishi",
    year = "2016",
    month = "12",
    day = "30",
    doi = "10.1109/HUMANOIDS.2016.7803258",
    language = "English",
    pages = "81--88",
    booktitle = "Humanoids 2016 - IEEE-RAS International Conference on Humanoid Robots",
    publisher = "IEEE Computer Society",
    address = "United States",

    }

    TY - GEN

    T1 - Material recognition CNNs and hierarchical planning for biped robot locomotion on slippery terrain

    AU - Brandão, Martim

    AU - Shiguematsu, Yukitoshi Minami

    AU - Hashimoto, Kenji

    AU - Takanishi, Atsuo

    PY - 2016/12/30

    Y1 - 2016/12/30

    N2 - In this paper we tackle the problem of visually predicting surface friction for environments with diverse surfaces, and integrating this knowledge into biped robot locomotion planning. The problem is essential for autonomous robot locomotion since diverse surfaces with varying friction abound in the real world, from wood to ceramic tiles, grass or ice, which may cause difficulties or huge energy costs for robot locomotion if not considered. We propose to estimate friction and its uncertainty from visual estimation of material classes using convolutional neural networks, together with probability distribution functions of friction associated with each material. We then robustly integrate the friction predictions into a hierarchical (footstep and full-body) planning method using chance constraints, and optimize the same trajectory costs at both levels of the planning method for consistency. Our solution achieves fully autonomous perception and locomotion on slippery terrain, which considers not only friction and its uncertainty, but also collision, stability and trajectory cost. We show promising friction prediction results in real pictures of outdoor scenarios, and planning experiments on a real robot facing surfaces with different friction.

    AB - In this paper we tackle the problem of visually predicting surface friction for environments with diverse surfaces, and integrating this knowledge into biped robot locomotion planning. The problem is essential for autonomous robot locomotion since diverse surfaces with varying friction abound in the real world, from wood to ceramic tiles, grass or ice, which may cause difficulties or huge energy costs for robot locomotion if not considered. We propose to estimate friction and its uncertainty from visual estimation of material classes using convolutional neural networks, together with probability distribution functions of friction associated with each material. We then robustly integrate the friction predictions into a hierarchical (footstep and full-body) planning method using chance constraints, and optimize the same trajectory costs at both levels of the planning method for consistency. Our solution achieves fully autonomous perception and locomotion on slippery terrain, which considers not only friction and its uncertainty, but also collision, stability and trajectory cost. We show promising friction prediction results in real pictures of outdoor scenarios, and planning experiments on a real robot facing surfaces with different friction.

    UR - http://www.scopus.com/inward/record.url?scp=85010208956&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85010208956&partnerID=8YFLogxK

    U2 - 10.1109/HUMANOIDS.2016.7803258

    DO - 10.1109/HUMANOIDS.2016.7803258

    M3 - Conference contribution

    AN - SCOPUS:85010208956

    SP - 81

    EP - 88

    BT - Humanoids 2016 - IEEE-RAS International Conference on Humanoid Robots

    PB - IEEE Computer Society

    ER -