Fig 1 - uploaded by Junzhi Yu
Content may be subject to copyright.
Source publication
This paper is concerned with the design of a robotic fish and its motion control algorithms. A radio-controlled, four-link biomimetic robotic fish is developed using a flexible posterior body and an oscillating foil as a propeller. The swimming speed of the robotic fish is adjusted by modulating joint's oscillating frequency, and its orientation is...
Contexts in source publication
Context 1
... large amplitude undu- lation is mainly confined to the last 1/3 part of the body, and thrust is produced by a rather stiff caudal fin. The amplitude of this undulation, however, is small, or zero, in the anterior portion of the fish, increasing drastically in the immediate vicinity of the trailing edge [26]. Based on this information, as shown in Fig. 1, a physical model of the carangiform motion can be divided into two parts: flexible body and oscillatory lunate caudal fin, where the flexible body is represented by a series of oscillatory hinge joints and the caudal fin by an oscillating foil. A relative swim- ming model for RoboTuna (carangiform) has been presented by Barrett et al. ...
Context 2
... the basis of hybrid control, as shown in Fig. 10, different strategies are chosen according to different distance between the fish body and the destination point. The measure being taken is from crude to fine. If , the fish speeds up to approach the destination; If , when the fish is in motion, ac- curate control is employed, that is, it slows down and approaches within a certain ...
Context 3
... verify the feasibility and reliability of the proposed al- gorithms, an experimental robotic fish system has been con- structed. The system, as depicted in Fig. 11, consists of four subsystems: the robotic fish subsystem, the vision subsystem, the decisions-making subsystem, and the communication sub- system. All aquatic experiments presented in this paper were carried out in a 2000 mm 1150 mm pond with still water. The information of the fishes and their surroundings captured by overhead CCD ...
Context 4
... overhead camera. By calling the steering function MoveToGoal continu- ously, where , the fish intentionally swam to- ward the ball, and sometime pushed it. Because the ball was too light to remain stationary, the fish lost it and pushed it again just like playing a game. This can be considered that the fish tracked the floating ball continuously. Fig. 12(a) shows a photo of an experimental scenario during playing-ball, Fig. 12(b) shows a moving trajectory of the fish swimming toward a ball, where the positions of the fish and the ball are denoted in image plane coordinates in which the whole view field is regarded as a plane with 320 240 pixels. Fig. 12(c) shows the corresponding ori- ...
Context 5
... where , the fish intentionally swam to- ward the ball, and sometime pushed it. Because the ball was too light to remain stationary, the fish lost it and pushed it again just like playing a game. This can be considered that the fish tracked the floating ball continuously. Fig. 12(a) shows a photo of an experimental scenario during playing-ball, Fig. 12(b) shows a moving trajectory of the fish swimming toward a ball, where the positions of the fish and the ball are denoted in image plane coordinates in which the whole view field is regarded as a plane with 320 240 pixels. Fig. 12(c) shows the corresponding ori- entation error . Notice that the pond is not large enough to remove the ...
Context 6
... tracked the floating ball continuously. Fig. 12(a) shows a photo of an experimental scenario during playing-ball, Fig. 12(b) shows a moving trajectory of the fish swimming toward a ball, where the positions of the fish and the ball are denoted in image plane coordinates in which the whole view field is regarded as a plane with 320 240 pixels. Fig. 12(c) shows the corresponding ori- entation error . Notice that the pond is not large enough to remove the effects of the reflective waves at present, so the po- sitions of the ball and the fish will be slightly varied with the disturbances. Notice also that the range of the orientation error toward the end of the experiment seems much ...
Context 7
... the fish seems to be un-steadier than it does at a steady speed . For these reasons, the actual orientation error is larger than the expected value. If the experiment is done with a larger pool, more confident results will be achieved. Experiment B: Passing a Hole: To test controllability of the robotic fish in a narrow space, as shown in Fig. 13, two bars marked with the predefined color are aligned in a line to form a "HOLE" with a clearance of 100 mm. For the robotic fish, its task was to pass through the hole from an arbitrary initial position and orientation. The fish status and the hole position (Hx,Hy) were located by the overhead ...
Context 8
... in the hole, namely, , let the fish swim with full-speed . After the fish passed through the hole, i.e., , the above men- tioned algorithm was repeated to make the fish pass from the other side of the bar again. Here, the constant denotes the length of the fish, and indi- cates the length of the forebody. To make it more intelligible, as shown in Fig. 13, a triangular safety area satisfying the requirements of and is defined. In this area, the fish moves straight and is in a stage of "on;" otherwise, the fish is in a stage of "off" and has to call the steering function to adjust itself till it enters into the "on" stage. The slim robotic fish, by means of simple continual "ON-OFF" ...
Context 9
... is in a stage of "on;" otherwise, the fish is in a stage of "off" and has to call the steering function to adjust itself till it enters into the "on" stage. The slim robotic fish, by means of simple continual "ON-OFF" control in obstacle avoidance, can successfully get across a narrow gap. An image sequence of passing the hole is demonstrated in Fig. 14. Compared to other obstacle-avoidance methods such as the potential field technology [33] and the distance transform method [34], where the robot is often represented as a point in configuration, our proposed triangular "on-off" control can be especially applied to the slim-shaped robot's collision-free. Of course, once the fish ...
Context 10
... Of course, once the fish integrated multiple various sensors is put into practice in future, more advanced local path-planning method considered the fish's shape will be further investigated. Discussions: By calling PTP control algorithm, at present, the fish can play with a ball and pass a hole in the pond. But, it is observed from Fig. 12(c) that the orientation error, between and 25 , does not appear to limit to zero. This implies that it has nonlinear motion, that is, the fish body oscillates contin- uously during moving. To some extent, the dynamic imbalance of gravitation and buoyancy at the flapping tail affects propul- sive performance and steadiness, which is ...
Citations
... Here, the simple integration process enables the PID controller to be adjusted while using nominal effectiveness and reliability as control parameters. [29] • The development of the robotic fish and its movement control strategies are the objectives of this study. • A versatile anterior body and a vibrating foil acting as propellers have been employed to create a remotecontrolled, four-linked analogous robotic fish. ...
Robots that can comprehend and navigate their surroundings independently on their own are considered intelligent mobile robots (MR). Using a sophisticated set of
controllers, artificial intelligence (AI), deep learning (DL), machine learning (ML), sensors, and computation for navigation, MR's can understand and navigate around their environments without even being connected to a cabled source of power. Mobility and intelligence are fundamental drivers of autonomous robots that are
intended for their planned operations. They are becoming popular in a variety of fields, including business, industry, healthcare, education, government, agriculture, military operations, and even domestic settings, to optimize everyday activities. We describe different controllers, including proportional integral derivative (PID) controllers, model predictive controllers (MPCs), fuzzy logic controllers (FLCs), and reinforcement learning controllers used in robotics science. The main objective of this article is to demonstrate a comprehensive idea and basic working principle of controllers utilized by mobile robots (MR) for navigation. This work thoroughly investigates several available books and literature to provide a better understanding of the navigation strategies taken by MR. Future research trends and possible challenges to optimizing the MR navigation system are also discussed.
... • Highlighting the significance of comprehending fish biomechanics for effective use. Yu et al [72] . ...
Over the past few years, the research community has witnessed a burgeoning interest in biomimetics, particularly within the marine sector. The study of biomimicry as a revolutionary remedy for numerous commercial and research-based marine businesses has been spurred by the difficulties presented by the harsh maritime environment. Biomimetic marine robots are at the forefront of this innovation by imitating various structures and behaviours of marine life and utilizing the evolutionary advantages and adaptations these marine organisms have developed over millennia to thrive in harsh conditions. This thorough examination explores current developments and research efforts in biomimetic marine robots based on their propulsion mechanisms. By examining these biomimetic designs, the review aims to solve the mysteries buried in the natural world and provide vital information for marine improvements. In addition to illuminating the complexities of these bio-inspired mechanisms, the investigation helps to steer future research directions and possible obstacles, spurring additional advancements in the field of biomimetic marine robotics. Considering the revolutionary potential of using nature's inventiveness to navigate and thrive in one of the most challenging environments on Earth, the conclusion of the current review urges a multidisciplinary approach by integrating robotics and biology. The field of biomimetic marine robotics not only represents a paradigm shift in our relationship with the oceans, it also opens previously unimaginable possibilities for sustainable exploration and use of marine resources by understanding and imitating nature's solutions.
... Research on bionics is primarily divided into structures and materials. In the field of underwater robotics, research related to structural bionics includes bionic fish [4,5], lobsters [6], turtles [7], manta [8], snakes [9], and other animals. The morphology, movement, and propulsion mechanisms of these animals are studied to design robots that are adapted to the underwater environment. ...
Underwater bionic-legged robots encounter significant challenges in attitude, velocity, and positional control due to lift and drag in water current environments, making it difficult to balance operational efficiency with motion stability. This study delves into the hydrodynamic properties of a bionic crab robot’s shell, drawing inspiration from the sea crab’s motion postures. It further refines the robot’s underwater locomotion strategy based on these insights. Initially, the research involved collecting attitude data from crabs during underwater movement through biological observation. Subsequently, hydrodynamic simulations and experimental validations of the bionic shell were conducted, examining the impact of attitude parameters on hydrodynamic performance. The findings reveal that the transverse angle predominantly influences lift and drag. Experiments in a test pool with a crab-like robot, altering transverse angles, demonstrated that increased transverse angles enhance the robot’s underwater walking efficiency, stability, and overall performance.
... For complex underwater environments, such as narrow cave exploration, biological population investigation, and mariculture monitoring, exceptional maneuverability is imperative for underwater vehicles (Ryuh et al., 2015). The agile biomimetic robotic fish, which emulates the exceptional swimming abilities of real fish, possesses significant value for research and development (Zhang et al., 2016;Yu et al., 2004). ...
Robotic fish have been extensively developed for scientific research and engineering applications, with the robotic fishtail serving as a crucial component. Due to the complex deformation mechanism and hydrodynamic interaction, it remains a significant challenge to establish a real-time precise kinematic and hydrodynamic model for flapping robotic fishtails. This paper presents a kinematic and hydrodynamic model for a wire-driven robotic fishtail, incorporating experimental investigation. The wire-driven mechanism featuring a continuously compliant backbone and position-variable vertebrae is developed, enabling more flexible fishtail flapping and adjustable fish morphology. The kinematic and hydrodynamic models are established through the quasi-static method with proposed corrections accounting for flapping velocity and nonlinear backbone deformation during a flapping stroke, precisely assessing real-time hydrodynamic responses. The proposed models have been validated through physical tests in an open-water environment, and exhibit superior response predictive ability compared to traditional methods. A significant force peak is observed at the beginning of a flapping stroke, while small amount of reversed thrust occurs at its end. In a high-frequency flapping motion, the soft caudal fin generates greater thrust than a hard fin. Comprehensively, this study presents a systematic research methodology for modeling and experimental investigation of novel robotic fishtails and biomimetic propellers.
... Fuzzy logic controllers (FLC) have been widely used to control dynamic systems exhibiting model uncertainties. In previous works, FLCs are used for speed [23], depth [24,25], and orientation [26] control which addresses the uncertainty model problems. Though these controllers have improved performance in unmodeled uncertainties, their effectiveness cannot always be guaranteed. ...
The development of bio-inspired aquatic robots for underwater operations and scientific research has dramatically improved over the past decades. Dynamic modelling of such robots relies on the reactive force produced through the physical movement of their body parts. It is highly complex to capture the complete hydrodynamics for defining the reactive force, making the modelling and control of robotic fish challenging. This paper captures the hydrodynamic parameters from real-time data using a grey-box model structure optimized by a genetic algorithm (GA). The GA-optimized model aligns with real-time experimentation, exhibiting an average mean square error of approximately 0.001 m² across six swimming speeds. Next, GA optimization is proposed for designing the fuzzy logic controller to control the speed of the robotic fish. GA adjusts the parameters of the membership function and minimizes the error function. Finally, the controller’s performance is compared with the classical GA-tuned PID (GAPID) and conventional fuzzy logic controller (FLC). The proposed controller has significant improvement in terms of tracking error, integral square error (ISE), integral absolute error and integral time absolute error. The optimized controller has achieved an ISE improvement of 84.64% and 87.15% compared to FLC and GAPID controllers, respectively.
... An attempt is made for a three-dimensional trajectory tracking problem in robotic fish using SMC via simulations. 14 Alternatively, Yu et al. 25 implemented PID and fuzzy controller for speed and heading control using visual feedback system and Wen et al. 26 designed a fuzzy logic controller for desired speed tracking. Comparing the simulation results of PID, fuzzy controllers produced better thrust efficiency. ...
This article proposes real-time speed tracking of two-link surface swimming robotic fish using a deep reinforcement learning (DRL) controller. Hydrodynamic modelling of robotic fish is done by virtue of Newtonian dynamics and Lighthill’s kinematic model. However, this includes external unsteady reactive forces that cannot be modeled accurately due to the distributed nature of hydrodynamic behavior. Therefore, a novel data-assisted dynamic model and control method is proposed for the speed tracking of robotic fish. Initially, the cruise speed motion data are collected through experiments. The water-resistance coefficient is estimated using the least mean square fit, which is then adopted in the model. Subsequently, a closed-loop discrete-time DRL controller trained through a soft actor-critic (SAC) agent is implemented through simulations. SAC overcomes the brittleness problem encountered by other policy gradient approaches by encouraging the policy network for maximum exploration and not assigning a higher probability to any single part of actions. Due to this robustness in the policy learning, the convergence error becomes low in RL-SAC than RL-DDPG controller. The simulation results verify that the DRL-SAC control with data-assisted modelling substantially improves the speed tracking performance. Further, this controller is validated in real-time, and it is observed that the SAC-trained controller tracks the desired speed more accurately than the DDPG controller.
... It is already being used for various tasks such as object identification, vessel hull inspections, and underwater survey missions [10][11][12]. The locomotion of fish has inspired researchers to design robots to mimic the motion and dynamics of natural fish, namely robotic fish, which is an Autonomous Underwater Vehicle (AUV) designed to obtain fish-like swimming behaviors [3,17,18]. Robotuna was the first-ever robotic fish prototype in the world, designed and developed by MIT in 1994. Since then, underwater robotics has reached new heights in achieving biomimicry in robotic fish [19]. ...
Numerous studies have been conducted to prove the calming and stress-reducing effects on humans of visiting aquatic environments. As a result, many institutions have utilized fish to provide entertainment and treat patients. The most common issue in this approach is controlling the movement of fish to facilitate human interaction. This study proposed an interactive robot, a robotic fish, to alter fish swarm behaviors by performing an effective, unobstructed, yet necessary, defined set of actions to enhance human interaction. The approach incorporated a minimalistic but futuristic physical design of the robotic fish with cameras and infrared (IR) sensors, and developed a fish-detecting and swarm pattern-recognizing algorithm. The fish-detecting algorithm was implemented using background subtraction and moving average algorithms with an accuracy of 78%, while the swarm pattern detection implemented with a Convolutional Neural Network (CNN) resulted in a 77.32% accuracy rate. By effectively controlling the behavior and swimming patterns of fish through the smooth movements of the robotic fish, we evaluated the success through repeated trials. Feedback from a randomly selected unbiased group of subjects revealed that the robotic fish improved human interaction with fish by using the proposed set of maneuvers and behavior.
... However, because the traveling wave equation can only be used to illustrate the Cruising-straight motion of fish, it is complicated to control the robotic fish to perform other motions. For example, when Yu et al. used this method to control the robotic fish, based on the Cruisingstraight control model, it is also necessary to apply a specific deflection command to each link of the robotic fish to make it turn in different ways [35][36][37]. It makes it difficult for researchers to obtain the precise gait of the robotic fish as a whole, which can lead to a decrease in the accuracy of gait control. ...
Fish locomotion which adopts body and/or caudal fin swimming mode consists of different motions, such as Cruisingstraight, Cruising-turn, and various fast turns, etc. Currently, there is no single mathematical model that could illustrate all these motions. Thus, for scientists and engineers, it is quite cumbersome and complicated to model and control different motions with multiple principles. In this paper, we proposed a general kinematic model to illustrate the kinematics of all aforementioned swimming motions. The model is synthesized by a nonlinear oscillator and a traveling wave equation. By changing four parameters extracted from the model, the kinematic model can demonstrate all the aforementioned swimming motions with different amplitudes and frequencies. To verify the model, we built a multi-joint robotic fish, and developed its dynamic model and control method to perform all the maneuvers under the guidance of the general kinematic model. Through this systematic methodology, one can easily study the principles of different swimming motions and design the multi-motions controller for a robotic fish through only one governing kinematic model
... Data is often collected by the microcontroller unit and transmitted to a computer via cable. 5. Control system: Using the feedback data provided by the sensors, an integrated closed-loop algorithm should be able to control the actuation mechanism for sustained swimming and complex maneuverings such as a C-turn (Yu et al., 2004;Feng et al., 2020). 6. Material selection: Rapid prototyping methods are desirable during the design phase. ...
In this study, we first identify seven criteria for the design of biomimetic robotic fish and focus on two of these tasks: shape and hydromechanics. For this purpose, fish body part measurements were obtained using previous work and also captured and harvested fish samples to construct computer-aided design (CAD) models of three fish species: Scomber scombrus, Sarda sarda, and Thunnus thynnus. Dead body drag and caudal fin flapping were considered as two separate problems. Computational simulations were carried out to determine drag and propulsive performance. Body drag simulations showed that the Reynolds dependence of the drag coefficient of three different species can be adequately expressed by a single laminar scaling correlation. At the same length and swimming speed, the Atlantic mackerel experiences the least drag. Caudal fin deformation simulations showed that the Atlantic bluefin tuna offers the highest thrust and efficiency. Peak efficiency is in the range of 31–35 percent observed at the same optimal Strouhal number, St = 0.5, for all species. It is shown that the aspect ratio as the main length scale influences propulsion performance.
... Fish or fish-inspired swimming are often studied using both experiments with biological [12], [13] or robotic fish [14], [15], or using computational simulations [16]- [18]. Experiments with robotic fish can directly reveal the swimming performance with the actual physics of robot-fluid interaction. ...
