Journal article Open Access

Model-Based Reinforcement Learning for Closed-Loop Dynamic Control of Soft Robotic Manipulators

Thuruthel, Thomas George; Falotico, Egidio; Renda, Federico; Laschi, Cecilia


JSON Export

{
  "files": [
    {
      "links": {
        "self": "https://zenodo.org/api/files/8216c77d-746a-41f6-9ef5-308c3067f695/thuruthel2018model.pdf"
      }, 
      "checksum": "md5:d2beb3db830b6d9801984f472e3b22ac", 
      "bucket": "8216c77d-746a-41f6-9ef5-308c3067f695", 
      "key": "thuruthel2018model.pdf", 
      "type": "pdf", 
      "size": 3478695
    }
  ], 
  "owners": [
    83206
  ], 
  "doi": "10.1109/TRO.2018.2878318", 
  "stats": {
    "version_unique_downloads": 113.0, 
    "unique_views": 21.0, 
    "views": 21.0, 
    "version_views": 21.0, 
    "unique_downloads": 113.0, 
    "version_unique_views": 21.0, 
    "volume": 441794265.0, 
    "version_downloads": 127.0, 
    "downloads": 127.0, 
    "version_volume": 441794265.0
  }, 
  "links": {
    "doi": "https://doi.org/10.1109/TRO.2018.2878318", 
    "latest_html": "https://zenodo.org/record/3759636", 
    "bucket": "https://zenodo.org/api/files/8216c77d-746a-41f6-9ef5-308c3067f695", 
    "badge": "https://zenodo.org/badge/doi/10.1109/TRO.2018.2878318.svg", 
    "html": "https://zenodo.org/record/3759636", 
    "latest": "https://zenodo.org/api/records/3759636"
  }, 
  "created": "2020-04-21T16:15:02.206746+00:00", 
  "updated": "2020-04-21T20:20:18.001265+00:00", 
  "conceptrecid": "3759635", 
  "revision": 2, 
  "id": 3759636, 
  "metadata": {
    "access_right_category": "success", 
    "doi": "10.1109/TRO.2018.2878318", 
    "description": "<p>Dynamic control of soft robotic manipulators is an open problem yet to be well explored and analyzed. Most of the current applications of soft robotic manipulators utilize static or quasi-dynamic controllers based on kinematic models or linearity in the joint space. However, such approaches are not truly exploiting the rich dynamics of a soft-bodied system. In this paper, we present a model-based policy learning algorithm for closed-loop predictive control of a soft robotic manipulator. The forward dynamic model is represented using a recurrent neural network. The closed-loop policy is derived using trajectory optimization and supervised learning. The approach is verified first on a simulated piecewise constant strain model of a cable driven under-actuated soft manipulator. Furthermore, we experimentally demonstrate on a soft pneumatically actuated manipulator how closed-loop control policies can be derived that can accommodate variable frequency control and unmodeled external loads.</p>", 
    "license": {
      "id": "CC-BY-4.0"
    }, 
    "title": "Model-Based Reinforcement Learning for Closed-Loop Dynamic Control of Soft Robotic Manipulators", 
    "relations": {
      "version": [
        {
          "count": 1, 
          "index": 0, 
          "parent": {
            "pid_type": "recid", 
            "pid_value": "3759635"
          }, 
          "is_last": true, 
          "last_child": {
            "pid_type": "recid", 
            "pid_value": "3759636"
          }
        }
      ]
    }, 
    "publication_date": "2018-11-12", 
    "creators": [
      {
        "affiliation": "BioRobotics Institute, Scuola Superiore Sant'Anna, Pisa, Italy", 
        "name": "Thuruthel, Thomas George"
      }, 
      {
        "affiliation": "BioRobotics Institute, Scuola Superiore Sant'Anna, Pisa, Italy", 
        "name": "Falotico, Egidio"
      }, 
      {
        "affiliation": "Department of Mechanical Engineering and the Center for Autonomous Robotics Systems, Khalifa University of Science and Technology, Abu Dhabi, United Arab Emirates", 
        "name": "Renda, Federico"
      }, 
      {
        "affiliation": "BioRobotics Institute, Scuola Superiore Sant'Anna, Pisa, Italy", 
        "name": "Laschi, Cecilia"
      }
    ], 
    "access_right": "open", 
    "resource_type": {
      "subtype": "article", 
      "type": "publication", 
      "title": "Journal article"
    }
  }
}
21
127
views
downloads
Views 21
Downloads 127
Data volume 441.8 MB
Unique views 21
Unique downloads 113

Share

Cite as