Modeling dynamic behavior of dielectric elastomer muscle for robotic applications

Area of Science:

• Materials Science
• Robotics
• Actuator Technology

Background:

• Dielectric elastomer actuators (DEAs) are promising artificial muscles due to their lightweight and large actuation capabilities.
• However, their practical application is hindered by non-linear responses, time-varying strain, and low load-bearing capacity stemming from their viscoelastic nature.
• Interactions between viscoelastic, dielectric, and conductive relaxations complicate performance prediction, especially in complex configurations like multilayer stacks.

Purpose of the Study:

• To review existing models for estimating the electro-mechanical response of DEAs.
• To propose a novel energy-based model integrating non-linear and time-dependent theories for predicting long-term DEA dynamic response.
• To validate the proposed model's accuracy against experimental data for future applications.

Main Methods:

• Review of established modeling strategies for DEAs.
• Development of a new hybrid model combining non-linear and time-dependent energy-based principles.
• Experimental validation of the model's predictive capability over extended durations (up to 20 minutes).

Main Results:

• The proposed model accurately predicts the long-term electro-mechanical dynamic response of DE muscles.
• The model demonstrated minimal errors when compared to experimental results over a 20-minute period.
• This enhanced predictive accuracy addresses key challenges in DEA performance estimation.

Conclusions:

• The developed model offers a robust solution for predicting the complex, long-term behavior of DE muscles.
• Accurate modeling is crucial for overcoming the limitations of DEAs in practical systems.
• Future work should focus on refining models for broader applications in robotics, haptics, and collaborative devices.