VLEIBot: A New 45-mg Swimming Microrobot Driven by a Bioinspired Anguilliform Propulsor

TL;DR

This work addresses autonomous underwater locomotion at millimeter-to-centimeter scales using simple onboard actuation. The authors introduce two microrobots, VLEIBot ($45$ mg) and VLEIBot+ ($90$ mg), whose tails undulate via a $6$-mg SMA actuator and a dual-propulsor configuration for 2D control, all leveraging fluid-structure interaction to generate propulsion. Tail-geometry optimization identifies a parabolic tail with $AR_p=0.41$ and length $26$ mm as optimal, achieving up to $15.1$ mm s$^{-1}$ for VLEIBot and $16.1$ mm s$^{-1}$ for VLEIBot+ at $5$ Hz, with VLEIBot+ providing turning rates up to $0.28$ rad s$^{-1}$ and RMS lateral tracking as low as $3.94$ mm in closed-loop. The results demonstrate self-propelled anguilliform-inspired microrobots capable of 2D navigation and cooperative operation, offering a platform for aquatic inspection and monitoring tasks in shallow water.

Abstract

This paper presents the VLEIBot^* (Very Little Eel-Inspired roBot), a 45-mg/23-mm^3 microrobotic swimmer that is propelled by a bioinspired anguilliform propulsor. The propulsor is excited by a single 6-mg high-work-density (HWD) microactuator and undulates periodically due to wave propagation phenomena generated by fluid-structure interaction (FSI) during swimming. The microactuator is composed of a carbon-fiber beam, which functions as a leaf spring, and shape-memory alloy (SMA) wires, which deform cyclically when excited periodically using Joule heating. The VLEIBot can swim at speeds as high as 15.1mm * s^{-1} (0.33 Bl * s^{-1}}) when driven with a heuristically-optimized propulsor. To improve maneuverability, we evolved the VLEIBot design into the 90-mg/47-mm^3 VLEIBot^+, which is driven by two propulsors and fully controllable in the two-dimensional (2D) space. The VLEIBot^+ can swim at speeds as high as 16.1mm * s^{-1} (0.35 Bl * s^{-1}), when driven with heuristically-optimized propulsors, and achieves turning rates as high as 0.28 rad * s^{-1}, when tracking path references. The measured root-mean-square (RMS) values of the tracking errors are as low as 4 mm.

Figures (7)

• Figure 1: Photograph of the VLEIBot and VLEIBot+. The VLEIBot (left) is a $45$-mg swimming microrobot driven by a new bioinspired propulsor. The VLEIBot+ (right) is a $90$-mg controllable microrobotic swimmer designed to achieve high maneuverability in the two-dimensional ($2$D) space.
• Figure 2: Robotic design and functionality.(a) Components and assembly of the VLEIBot. The VLEIBot is composed of four main components: (i) a rigid body that supports the weight of the robot on water due to surface tension; (ii) a $6$-mg HWD SMA-based actuator with passive hinges installed at its two distal ends; (iii)~a planar four-bar transmission mechanism that maps the output displacement generated by the SMA-based actuator into the angular oscillation that excites the undulating tail of the swimmer; and, (iv)~an anguilliform-inspired soft tail made of fluoropolymer film. The triplet $\left\{ \boldsymbol{n}_1,\boldsymbol{n}_2,\boldsymbol{n}_3 \right\}$ denotes the inertial frame used to describe the kinematics of the system. (b) Transmission functionality. The planar four-bar mechanism can be adjusted to have a constant bias angle, $\phi_\text{b}$, by shifting the installation point of the fixed end by a distance $d_\text{b}$. (c) Forward locomotion. The VLEIBot is designed to function with a symmetric stroke envelope (sweeping area). The SMA-based actuator produces the periodic displacement output that is mapped by the transmission into the large stroke angles required to undulate the robot's tail. The triplet $\left\{ \boldsymbol{b}_1,\boldsymbol{b}_2,\boldsymbol{b}_3 \right\}$ denotes the body-fixed frame used to describe the kinematics of the system. (d) Right turn. Theoretically, a right-biased undulation produces a negative torque with respect to the body frame and, as a consequence, a right turn. (e) Left turn. Theoretically, a left-biased undulation produces a positive torque with respect to the body frame and, as a consequence, a left turn.
• Figure 3: Experimental setup used in the tail-characterization swimming tests.(a) Experimental setup. A Simulink Real-Time host-target system and an AD/DA board (National Instruments PCI-$6229$) are used to generate the PWM signal that excites the robot. This signal is then filtered through a waveform amplifier (Accel Instruments TS$250$-$02$) to provide the power necessary to periodically Joule heat the SMA-based actuator that drives the VLEIBot. During the swimming tests, the VLEIBot is placed in a pool filled with water and a four-VK16-camera Vicon motion-capture system is used to measure the instantaneous position and orientation of the robot at a rate of $250\,\text{Hz}$. (b) Parameters of the parabolic tails with constant aspect ratio. For the tail-characterization experiments, we kept the aspect ratio constant at $\AR_{\text{p}} = 0.41$ and varied the tail length, $l_{\text{p}}$, in increments of $2\,\text{mm}$ over the range $\left[0:28\right]\,\text{mm}$. The height $h_{\text{p}}$ directly depends on the length $l_{\text{p}}$. (c) Parameters of the rectangular tails with constant height. For the tail-characterization experiments, we kept the height, $h_\text{r}$, constant at $4\,\text{mm}$ and varied the length, $l_\text{r}$, in increments of $5\,\text{mm}$ over the range $[0:50]\,\text{mm}$.
• Figure 4: Tail-characterization data.(a) Characterization of parabolic tails with constant aspect ratio. During the swimming tests, the aspect ratio was kept constant at $\AR_{\text{p}} = 0.41$, while we investigated the relationship between tail length and average forward speed across frequencies of $1$, $5$, $10$, $15$, and $20$ Hz. We see an increasing trend in average speed as tail length increases. At the tail length of $26\,\text{mm}$ and frequency of $5\,\text{Hz}$, we measured the maximum average speed of $15.1\,\text{mm} \cdot \text{s}^{-1}$ ($0.33\,\text{Bl} \cdot \text{s}^{-1}$). When the tail length, $l_{\text{p}}$, was increased further, the weight of the tail tended to drag the robot under the surface of the water. (b) Characterization of rectangular tails with constant height. During the swimming tests, the tail height was kept constant at $h_{\text{r}} = 4\,\text{mm}$, while we investigated the relationship between tail length and speed. We varied the tail length in increments of $5\,\text{mm}$ over the range $\left[5:50\right]\,\text{mm}$. It can be observed that high-frequency actuation is not able to produce thrust with any of the investigated tails, and $1\,\text{Hz}$ is the best actuation frequency for all the tested profiles. At the length of $20\,\text{mm}$, we see the best swimming performance of $4.7\,\text{mm} \cdot \text{s}^{-1}$ ($0.12\,\text{Bl} \cdot \text{s}^{-1}$).
• Figure 5: Locomotion experiments performed using optimized parabolic $\boldsymbol{26}$-mm tail with constant aspect ratio of $\boldsymbol{0.41}$.(a) Photographic composite from video footage showing the VLEIBot swimming at $1\,\text{Hz}$. The duty cycle of the driving PWM signal is $8\,\%$. (b) Photographic composite from video footage showing the VLEIBot swimming at $5\,\text{Hz}$. The duty cycle of the driving PWM signal is $12\,\%$. (c) Photographic composite from video footage showing the VLEIBot+ swimming at $1$ Hz. The duty cycles of the driving PWM signals are $8\,\%$. (d) Photographic composite from video footage showing the VLEIBot+ swimming at $5\,\text{Hz}$. The duty cycles of the driving PWM signals are $12\,\%$. Video footage of these experiments can be viewed in the accompanying supplementary movie.