Capturing Motion,
Crafting Stories

Recently, the latest achievement of the research team led by Professor Zhou Yitong from South China University of Technology, "Manta Ray-Inspired Soft Robotic Swimmer for High-speed and Multi-modal Swimming," was accepted by the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024).

The research team proposed a novel soft robotic swimmer inspired by manta rays, featuring bistable flapping wings driven by McKibben artificial muscles, with a maximum swimming speed of 12.23 cm per second and a maximum turning angular velocity of 22.5 degrees per second, while also achieving multi-modal swimming including forward and backward translation, turning, and flip-turning.

NOKOV Motion Capture System provided real-time speed data for the robots swimming, recording its motion state under different driving conditions, aiding in the optimization of the robots performance and design.

Citation

Z. Xu, J. Liang, Y. Zhou,”Manta Ray-Inspired Soft Robotic Swimmer for High-speed and Multi-modal Swimming,” in 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024).

Background

Manta rays can swim quickly and efficiently through flexible flapping wings, with a propulsion efficiency of up to 89%. Inspired by this, researchers have developed multiple soft robotic swimmers using the flapping wing principle for ecological monitoring, deep-sea exploration, and aquaculture inspection. However, limited by the low stiffness and viscosity of soft materials, these robots are confined to low speeds and simple swimming modes. To develop fast and efficient soft robotic swimmers, Professor Zhou Yitongs research team proposed a bistable wing design based on mechanically prestressed composite materials, capable of high-speed movement while achieving various swimming modes, providing new technical support for fields such as underwater exploration and environmental monitoring.

Experimental Process

The research team created bistable flapping wings with a simple structure, fast manufacturing speed, and customizable morphology by pre-stretching the crossed elastomer layers on both sides of a PET thin plate. Lightweight McKibben artificial muscles are integrated on both sides of the wings to drive the snap-through of the wings, which have the advantages of low energy consumption, small size, and minimal influence from the underwater environment compared to shape memory alloys and dielectric elastomer actuators. By controlling the pressure applied to different McKibben actuators, multimodal swimming can be achieved, including forward and backward translation, turning, and flipping.

Design and bistable shapes of the soft bistable manta ray-inspired robot. (a) Structural design of the robot. (b) The robots two stable states I and II. In state II, pneumatic actuation of McKibben artificial muscles (A) (P2>0) triggers the wing to snap to stable state I, and vice versa, V represents the swimming direction.

An experimental pneumatic platform was set up to provide the robot with different pressures and frequencies of driving, and the NOKOV Motion Capture System recorded the robots motion speed under different driving conditions, including translational speed and rotational angular velocity, conducting various modal demonstration experiments.

The results showed that the proposed swimming robot has multimodal swimming capabilities, achieving a maximum translational speed of 12.23 cm/s and a turning speed of 22.5°/s in bistable mode, more than twice that of the monostable mode. This study provides a new paradigm for designing manta ray-like robots with high-speed and multimodal swimming capabilities. Future exploration directions include the unconstrained design of swimming robots and further improving the robots swimming energy efficiency compared to the actual efficiency of manta rays.

NOKOV Motion Capture System provided real-time speed data for the robots swimming, recording its motion state under different driving conditions, aiding in the optimization of the robots performance and design.

Author Introduction

Zefeng Xu, Ph.D. student at South China University of Technology. Main research directions include robotic mechanics modeling, intelligent control technology. As the first author, he has published 4 SCI papers in journals such as RAL and JBE, 3 papers in international EI conferences.He was a finalist for the best paper award at the 2022 International Conference on Robotics and Biomimetics (IEEE ROBIO).

Jiaqiao Liang, Master student at South China University of Technology. Main research directions include underwater robots, soft robots, control strategy design.

Yitong Zhou (Corresponding Author), Associate Professor and Ph.D. supervisor at the Shien-Ming Wu School of Intelligent Engineering, South China University of Technology. She was a finalist for the best paper award at the 2022 International Conference on Robotics and Biomimetics as a mentor, and the 2022 Journal of Mechanisms and Robotics best paper award. She has published 24 research papers in well-known journals and conferences such as IEEE TRO / IEEE TIM / IEEE Sensors / ASME JMR / RAL/ ICRA / IROS.