Skip to content

Towards efficient elastic actuation in bio-inspired robotics using dielectric elastomer artificial muscles

Research output: Contribution to journalArticle

Original languageEnglish
Article number095015
Number of pages16
JournalSmart Materials and Structures
Issue number9
Early online date15 Jul 2019
DateAccepted/In press - 15 Jul 2019
DateE-pub ahead of print - 15 Jul 2019
DatePublished (current) - 12 Aug 2019


In nature, animals reduce their cost of transport by utilizing elastic energy recovery. Emerging soft robotic technologies such as dielectric elastomer actuators (DEAs) offer an advantage in achieving biomimetic energy efficient locomotion thanks to their high actuation strain and inherent elasticity. In this work, we conduct a comprehensive study on the feasibility of using antagonistic DEA artificial muscles for bio-inspired robotics. We adopt a double cone DEA configuration and develop a mathematical model to characterize its dynamic electromechanical response. It is demonstrated that this DEA design can be optimized in terms of the maximum work output by adjusting the strut height design parameter. Using this optimized design, we analyse the power/stroke output and the electromechanical efficiency of the DEA and show how these actuation characteristics can be maximized for different payload conditions, excitation frequencies and actuation waveforms. The elastic energy recovery from the DEA is then demonstrated by reducing the duty ratio of the actuation signal and thus allowing the stored elastic energy in the DEA membranes to contribute to the work output. A bio-inspired three-segment leg prototype driven by the same actuator is presented to demonstrate that the same energy recovery principle is feasible for bio-inspired robotics.

    Research areas

  • dielectric elastomer actuator, energetic study, electromechanical model, elastic energy recovery, bio-inspired locomotion, soft robotics



  • Full-text PDF (Author’s accepted manuscript)

    Rights statement: This is the author accepted manuscript (AAM). The final published version (version of record) is available online via IOP Publishing at . Please refer to any applicable terms of use of the publisher.

    Accepted author manuscript, 1.66 MB, PDF document

    Embargo ends: 15/07/20

    Request copy

    Licence: CC BY-NC-ND


View research connections

Related faculties, schools or groups