Bayesian-Optimized Impedance Control of an Aerial Robot for Safe Physical Interaction with the Environment

Asem Khattab, Ramy Rashad, Johan Engelen, Stefano Stramigioli

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    23 Downloads (Pure)

    Abstract

    Impedance control is a widely used interaction-control technique for aerial and ground robots. To achieve consistent performance during impedance control tasks, an a-priori knowledge of the environment parameters is needed to adjust the controller's impedance parameters accordingly. Concentrating on tasks requiring constant impedance parameters throughout operation, a model-free learning framework is proposed to autonomously find the suitable parameters values. The framework relies on Bayesian optimization and episodic reward calculation requiring the drone to repeatedly perform a predetermined task in the environment actively searching in the impedance parameters space. The sample-efficiency and safety of learning were improved by adding two novel modifications to standard Bayesian optimization. The proposed technique was validated in a high fidelity simulation environment. The results show that the proposed framework is able to automatically find suitable impedance parameters values in different situations given the same initial knowledge and that the learned parameters values can be generalized to similar interaction tasks.
    Original languageEnglish
    Title of host publication2019 IEEE International Symposium on Safety, Security, and Rescue Robotics, SSRR 2019
    PublisherIEEE
    Pages172-179
    Number of pages8
    ISBN (Electronic)9781728107783
    DOIs
    Publication statusPublished - 2 Sep 2019
    Event2019 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR) - University of Würzburg Graduate Schools building, Würzburg, Germany
    Duration: 2 Sep 20194 Sep 2019
    http://ssrr2019.org

    Conference

    Conference2019 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)
    Abbreviated titleSSRR 2019
    CountryGermany
    CityWürzburg
    Period2/09/194/09/19
    Internet address

    Fingerprint

    Robots
    Antennas
    Antenna grounds
    Controllers
    Drones

    Cite this

    Khattab, A., Rashad, R., Engelen, J., & Stramigioli, S. (2019). Bayesian-Optimized Impedance Control of an Aerial Robot for Safe Physical Interaction with the Environment. In 2019 IEEE International Symposium on Safety, Security, and Rescue Robotics, SSRR 2019 (pp. 172-179). [8848967] IEEE. https://doi.org/10.1109/SSRR.2019.8848967
    Khattab, Asem ; Rashad, Ramy ; Engelen, Johan ; Stramigioli, Stefano . / Bayesian-Optimized Impedance Control of an Aerial Robot for Safe Physical Interaction with the Environment. 2019 IEEE International Symposium on Safety, Security, and Rescue Robotics, SSRR 2019. IEEE, 2019. pp. 172-179
    @inproceedings{b383604f40f545de915f16aa289bfb82,
    title = "Bayesian-Optimized Impedance Control of an Aerial Robot for Safe Physical Interaction with the Environment",
    abstract = "Impedance control is a widely used interaction-control technique for aerial and ground robots. To achieve consistent performance during impedance control tasks, an a-priori knowledge of the environment parameters is needed to adjust the controller's impedance parameters accordingly. Concentrating on tasks requiring constant impedance parameters throughout operation, a model-free learning framework is proposed to autonomously find the suitable parameters values. The framework relies on Bayesian optimization and episodic reward calculation requiring the drone to repeatedly perform a predetermined task in the environment actively searching in the impedance parameters space. The sample-efficiency and safety of learning were improved by adding two novel modifications to standard Bayesian optimization. The proposed technique was validated in a high fidelity simulation environment. The results show that the proposed framework is able to automatically find suitable impedance parameters values in different situations given the same initial knowledge and that the learned parameters values can be generalized to similar interaction tasks.",
    author = "Asem Khattab and Ramy Rashad and Johan Engelen and Stefano Stramigioli",
    year = "2019",
    month = "9",
    day = "2",
    doi = "10.1109/SSRR.2019.8848967",
    language = "English",
    pages = "172--179",
    booktitle = "2019 IEEE International Symposium on Safety, Security, and Rescue Robotics, SSRR 2019",
    publisher = "IEEE",
    address = "United States",

    }

    Khattab, A, Rashad, R, Engelen, J & Stramigioli, S 2019, Bayesian-Optimized Impedance Control of an Aerial Robot for Safe Physical Interaction with the Environment. in 2019 IEEE International Symposium on Safety, Security, and Rescue Robotics, SSRR 2019., 8848967, IEEE, pp. 172-179, 2019 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), Würzburg, Germany, 2/09/19. https://doi.org/10.1109/SSRR.2019.8848967

    Bayesian-Optimized Impedance Control of an Aerial Robot for Safe Physical Interaction with the Environment. / Khattab, Asem; Rashad, Ramy; Engelen, Johan; Stramigioli, Stefano .

    2019 IEEE International Symposium on Safety, Security, and Rescue Robotics, SSRR 2019. IEEE, 2019. p. 172-179 8848967.

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    TY - GEN

    T1 - Bayesian-Optimized Impedance Control of an Aerial Robot for Safe Physical Interaction with the Environment

    AU - Khattab, Asem

    AU - Rashad, Ramy

    AU - Engelen, Johan

    AU - Stramigioli, Stefano

    PY - 2019/9/2

    Y1 - 2019/9/2

    N2 - Impedance control is a widely used interaction-control technique for aerial and ground robots. To achieve consistent performance during impedance control tasks, an a-priori knowledge of the environment parameters is needed to adjust the controller's impedance parameters accordingly. Concentrating on tasks requiring constant impedance parameters throughout operation, a model-free learning framework is proposed to autonomously find the suitable parameters values. The framework relies on Bayesian optimization and episodic reward calculation requiring the drone to repeatedly perform a predetermined task in the environment actively searching in the impedance parameters space. The sample-efficiency and safety of learning were improved by adding two novel modifications to standard Bayesian optimization. The proposed technique was validated in a high fidelity simulation environment. The results show that the proposed framework is able to automatically find suitable impedance parameters values in different situations given the same initial knowledge and that the learned parameters values can be generalized to similar interaction tasks.

    AB - Impedance control is a widely used interaction-control technique for aerial and ground robots. To achieve consistent performance during impedance control tasks, an a-priori knowledge of the environment parameters is needed to adjust the controller's impedance parameters accordingly. Concentrating on tasks requiring constant impedance parameters throughout operation, a model-free learning framework is proposed to autonomously find the suitable parameters values. The framework relies on Bayesian optimization and episodic reward calculation requiring the drone to repeatedly perform a predetermined task in the environment actively searching in the impedance parameters space. The sample-efficiency and safety of learning were improved by adding two novel modifications to standard Bayesian optimization. The proposed technique was validated in a high fidelity simulation environment. The results show that the proposed framework is able to automatically find suitable impedance parameters values in different situations given the same initial knowledge and that the learned parameters values can be generalized to similar interaction tasks.

    U2 - 10.1109/SSRR.2019.8848967

    DO - 10.1109/SSRR.2019.8848967

    M3 - Conference contribution

    SP - 172

    EP - 179

    BT - 2019 IEEE International Symposium on Safety, Security, and Rescue Robotics, SSRR 2019

    PB - IEEE

    ER -

    Khattab A, Rashad R, Engelen J, Stramigioli S. Bayesian-Optimized Impedance Control of an Aerial Robot for Safe Physical Interaction with the Environment. In 2019 IEEE International Symposium on Safety, Security, and Rescue Robotics, SSRR 2019. IEEE. 2019. p. 172-179. 8848967 https://doi.org/10.1109/SSRR.2019.8848967