*2.3.3. Free-swimming test*

The robot was tested in a water tank (1.5 m wide, 4.7 m long, and 0.9 m deep). The robot was positively buoyant. The operating frequency of the square wave actuation voltage was tuned to 0.167 Hz and the amplitude was set to 4 V. Since each pectoral fin consumed less than 1.2 W (shown in Fig. 11), the overall power consumption of the robot was under 2.5 W. A digital video camera (VIXIA HG21, Canon) was used to capture the motion of the swimming robot. Fig. 16 presents six snap shots of the swimming robot from top view, where the trajectories of left and right fins are plotted in blue and red, respectively.

**Figure 16.** Snap shots of free-swimming robotic manta ray.

Each snap shot was taken every 5 seconds. A swim speed of 0.74 cm/s was calculated from the movie using the Edge Detection program in the LabVIEW. Since the body length was 11 cm, the speed in body length per second (BL/s) was 0.067.

## **2.4. Comparison of ray-like swimming robots**

In our previous work, we developed an artificial pectoral fin that consists of four IPMCs bonded by PDMS membrane. The robot swam at 0.42 cm/s (0.053 BL/s) consuming 1 W in power. Many other research groups have been developed ray-like robots, using IPMCs, servomotors, and shape memory alloys as actuators. The comparison of ray-like swimming robots is shown in Table 1. It indicates that the IPMC powered ray-like robots are lighter and consume less power than the robots actuated by a servomotor or SMA. But they swim at lower speed for the reasons that IPMC is unable to flap at high frequency and generated force is also very low.


Ionic Polymer-Metal Composite Artificial Muscles in Bio-Inspired Engineering Research: Underwater Propulsion 237

**Table 1.** Comparison of Ray-like Robots.

236 Smart Actuation and Sensing Systems – Recent Advances and Future Challenges

**Figure 16.** Snap shots of free-swimming robotic manta ray.

**2.4. Comparison of ray-like swimming robots** 

force is also very low.

cm, the speed in body length per second (BL/s) was 0.067.

The robot was tested in a water tank (1.5 m wide, 4.7 m long, and 0.9 m deep). The robot was positively buoyant. The operating frequency of the square wave actuation voltage was tuned to 0.167 Hz and the amplitude was set to 4 V. Since each pectoral fin consumed less than 1.2 W (shown in Fig. 11), the overall power consumption of the robot was under 2.5 W. A digital video camera (VIXIA HG21, Canon) was used to capture the motion of the swimming robot. Fig. 16 presents six snap shots of the swimming robot from top view,

Each snap shot was taken every 5 seconds. A swim speed of 0.74 cm/s was calculated from the movie using the Edge Detection program in the LabVIEW. Since the body length was 11

In our previous work, we developed an artificial pectoral fin that consists of four IPMCs bonded by PDMS membrane. The robot swam at 0.42 cm/s (0.053 BL/s) consuming 1 W in power. Many other research groups have been developed ray-like robots, using IPMCs, servomotors, and shape memory alloys as actuators. The comparison of ray-like swimming robots is shown in Table 1. It indicates that the IPMC powered ray-like robots are lighter and consume less power than the robots actuated by a servomotor or SMA. But they swim at lower speed for the reasons that IPMC is unable to flap at high frequency and generated

where the trajectories of left and right fins are plotted in blue and red, respectively.

*2.3.3. Free-swimming test* 

The advantages of using IPMCs as artificial muscle in ray-like swimming robots are: 1) low actuation voltage; 2) works well in wet conditions; 3) no gears and motors, 4) simple mechanical structure; 5) low noise; 6) able to be shaped in bio-inspired engineering design. However, IPMC can only generate low force with slow response time, which limits the swimming speed of the robot. To accommodate the disadvantages and utilize the advantages, an optimal design of the pectoral fin must be conducted, where the dimensions and location of the IPMC actuator are optimized. The challenge arises from modelling the IPMC actuating membrane to produce optimal 3D kinematic motions, which is a future focus. Another focus will be on modelling of the fluid dynamics introduced by the 3D kinematic motions of the fin, to understand how the thrust force is generated.
