Development of the bioinspired propulsion system for a robotic fish

Abstract

Biomimetic robots aim to replicate the movement principles of living creatures, which have been continuously perfected by nature to ensure survival. Technically, survival comprises two factors: optimizing movement efficacy, such as speed, distance traveled, or acceleration, and minimizing energy consumption during movement. Researchers have made several attempts to develop a biomimetic fish robot [1–3]. Tunas are a group of fish with numerous evolutionary adaptations that make them exceptional swimmers, including tail shape, lateral peduncle keels, pectoral fin shape, and finlets. Thunniform locomotion is characterized by limited undulation, typically restricted to the rear one-third of the body, with maximal amplitude reached at the end of the tail peduncle. A bioinspired propulsion system was developed for a robotic fish model. It is based on the combination of an elastic cord with a tail fin that is firmly attached to the cord. Two symmetric movable thrusts that simulate muscle contractions connect the tail fin to a servomotor. The propulsion system provides oscillatory tail movement that can be controlled for amplitude and frequency. This movement translates to the movement of the robotic fish, which executes the thunniform principle of locomotion. The body and tail fin of the robotic fish were designed using a computational model that simulates a virtual body in water. Subsequently, we constructed a prototype of the robotic fish and tested it under experimental conditions. Experiments were conducted to investigate the relationship between the robot’s kinematics and the dynamic parameters of the propulsion system. The results showed that increasing the frequency of tail fin oscillations led to an increase in the robot’s speed. At fixed frequencies, there was an interval of energetically efficient travel speeds up to a threshold velocity. Movement at higher speeds was achievable; however, it was accompanied by greater power consumption. The conclusion aligns with the data from COT studies on living tunas, indicating qualitative agreement. However, the robot’s values were quantitatively higher. Our robotic fish reached a maximum speed of approximately 0.4 BL/s, exceeding speeds in other works where a simplified tail section was used (0.22 BL/s [4], 0.254 BL/s [5]). Our study demonstrated a correlation between the efficiency of robot swimming and amplitude, which was previously unexplored in addition to previous findings on the relationship between tail beat frequency and swimming speed, as well as the dependence of frequency and swimming speed. The results showed that increasing the oscillation amplitude of the propulsion system, at a fixed frequency, only led to a rise in swimming speed up to a certain threshold. However, further increases in amplitude resulted in minimal speed increases at higher energy costs. An evaluation of energy efficiency was conducted, assessing its dependence on dynamic parameters of tail oscillation. The results show that the transport cost rises as the tail amplitude rises beyond the threshold. Furthermore, it was found that for a fixed frequency, an interval of energetic preferred speeds up to a threshold exists. Although moving at a higher speed is possible, it consumes more power. In general, it is preferable to increase caudal fin oscillation frequency rather than amplitude to increase swimming speed. These findings are in qualitative accordance with the outcomes of the numerical simulation of thunniform swimming.