Undulatory Swimming Performance Explored With a Biorobotic Fish and Measured by Soft Sensors and Particle Image Velocimetry

Fabian Schwab, Fabian Wiesemüller, Claudio Mucignat, Yong‐Lae Park, Ivan Lunati, Mirko Kovač, Ardian JusufiFrontiers in Robotics and AI 2022 · DOI

TL;DR

This paper introduces a soft robotic fish equipped with integrated soft strain sensors to study undulatory swimming. By combining Particle Image Velocimetry (PIV) and Dynamic Mode Decomposition (DMD), they characterize the robot's performance across various flow conditions and gain insights into the underlying fluid dynamics, offering a new platform for bio-inspired robotics research.

What Problem Does This Paper Address?

Studying the intricate mechanics of fish swimming in their natural environment is challenging due to the difficulty of precisely controlling and measuring live animals. This limits our understanding of how fish achieve efficient locomotion, particularly how they adapt their body kinematics and muscle activation to varying water conditions. To bridge this gap, researchers are turning to bio-inspired soft robots as experimental platforms. This work builds upon previous soft robotic fish designs by integrating advanced soft sensing and sophisticated flow analysis techniques to provide a more comprehensive understanding of undulatory swimming.

Methods

The core of the research is a soft robotic fish constructed from silicone actuators and a flexible fin. A key innovation is the integration of eutectic gallium-indium (eGaIn) liquid metal strain sensors directly into the robot's structure. These sensors allow for real-time measurement of the fish's lateral deflection, providing crucial kinematic data.

To control the robot's swimming, a PID controller was implemented to maintain a constant undulatory amplitude, even when faced with varying flow speeds. This controller's effectiveness was tested in a recirculating flow tank.

For analyzing the swimming performance and the resulting fluid dynamics, the researchers employed Particle Image Velocimetry (PIV). This technique visualizes the flow field around the robot by tracking seeded particles. To further dissect the complex flow patterns, Dynamic Mode Decomposition (DMD) was used. DMD decomposes the flow field into dominant spatio-temporal modes, offering insights into the underlying physics and efficiency of the swimming motion.

Datasets

Not discussed in the paper. The paper focuses on experimental data collected from the robotic platform in a controlled flow tank environment.

Key Insights and Ideas

  • Soft Sensors as a Game Changer: The integration of eGaIn soft strain sensors provides a non-intrusive, real-time method to measure the robot's body deflection, crucial for understanding undulatory motion without external attachments.
  • Adaptive Control for Robustness: The PID controller's ability to maintain a constant tail amplitude across different flow speeds demonstrates a significant step towards mimicking the adaptability of real fish.
  • PIV and DMD Unveil Flow Secrets: Combining PIV for flow visualization with DMD for mode decomposition offers a powerful framework to analyze the complex wake structures generated by the robotic fish, revealing insights into thrust generation and efficiency.
  • Robophysics as a Biological Tool: The paper highlights how soft robotic platforms can serve as "model animals" for biomechanics, allowing for controlled experiments that are impossible with live specimens, thus advancing both robotics and biological understanding.

Key Results

  • The eGaIn sensors demonstrated a linear response to bending with a high R² value (0.952) and a low relative error, proving their reliability for measuring deflection.
  • The closed-loop controller successfully maintained a constant tail amplitude across a range of undulation frequencies (0.3-0.7 Hz) and flow speeds (0-19.4 cm/s), outperforming a feed-forward control approach.
  • PIV measurements revealed distinct V-shaped jet patterns in the wake, with the angle of these jets varying with undulation frequency and flow speed.
  • The Wake Power Model, derived from PIV data, provided a qualitative measure of thrust, suggesting a maximum self-propelled speed around 15 cm/s and an optimal actuation frequency of 0.6 Hz.
  • DMD analysis identified dominant flow modes, including one directly related to the tail's oscillatory motion and others linked to vortex shedding at multiples of the undulation frequency.

Limitations and Future Work

  • 2D Flow Approximation: The current analysis relies on 2D PIV measurements, which are a simplification of the inherently 3D flow field around a swimming robot. More accurate estimates of wake power and thrust would require 3D PIV.
  • Single Actuator Pair: The current design uses a single pair of soft actuators. Future work could involve integrating multiple actuators to achieve more complex body kinematics.
  • Sensor Durability: While the sensors showed good performance, the paper mentions enhancing sensor features like durability and flexibility.
  • Future Directions: The authors envision using 3D-PIV with DMD as a universal framework for comparing various bio-inspired swimming robots with real aquatic animals. They also plan to couple 3D velocity measurements with pressure field/thrust reconstruction and DMD to further investigate the influence of geometry, material properties, and control strategies on swimming performance. Enhancing the interface between soft actuators and sensors is also a key future goal.

Summary updated

Abstract

Due to the difficulty of manipulating muscle activation in live, freely swimming fish, a thorough examination of the body kinematics, propulsive performance, and muscle activity patterns in fish during undulatory swimming motion has not been conducted. We propose to use soft robotic model animals as experimental platforms to address biomechanics questions and acquire understanding into subcarangiform fish swimming behavior. We extend previous research on a bio-inspired soft robotic fish equipped with two pneumatic actuators and soft strain sensors to investigate swimming performance in undulation frequencies between 0.3 and 0.7 Hz and flow rates ranging from 0 to 20 <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mfrac><mml:mrow><mml:mi>c</mml:mi> <mml:mi>m</mml:mi></mml:mrow> <mml:mrow><mml:mi>s</mml:mi></mml:mrow> </mml:mfrac> </mml:math> in a recirculating flow tank. We demonstrate the potential of eutectic gallium-indium (eGaIn) sensors to measure the lateral deflection of a robotic fish in real time, a controller that is able to keep a constant undulatory amplitude in varying flow conditions, as well as using Particle Image Velocimetry (PIV) to characterizing swimming performance across a range of flow speeds and give a qualitative measurement of thrust force exerted by the physical platform without the need of externally attached force sensors. A detailed wake structure was then analyzed with Dynamic Mode Decomposition (DMD) to highlight different wave modes present in the robot's swimming motion and provide insights into the efficiency of the robotic swimmer. In the future, we anticipate 3D-PIV with DMD serving as a global framework for comparing the performance of diverse bio-inspired swimming robots against a variety of swimming animals.
BibTeX
@inproceedings{schwab2022undulatory,
  author = {Fabian Schwab and Fabian Wiesemüller and Claudio Mucignat and Yong‐Lae Park and Ivan Lunati and Mirko Kovač and Ardian Jusufi},
  title = {{Undulatory Swimming Performance Explored With a Biorobotic Fish and Measured by Soft Sensors and Particle Image Velocimetry}},
  booktitle = {Frontiers in Robotics and AI},
  year = {2022},
  doi = {10.3389/frobt.2021.791722},
}
Figure