In the field of robotics, achieving seamless operation across diverse environments remains a significant challenge. Traditional robots are often constrained by their working conditions; for instance, underwater robots typically lack terrestrial mobility, necessitating transportation to specific水域 for deployment and retrieval after tasks. To address these limitations, we turn to biomimicry, drawing inspiration from animals that naturally navigate both aquatic and terrestrial realms. Among these, the sea turtle stands out for its efficient flapping propulsion in water and adaptable crawling on land. This paper presents the design and analysis of an amphibious bionic robot inspired by the sea turtle’s locomotion mechanisms. We focus on developing a bionic robot that integrates flapping mechanisms, deformable self-locking hydrofoils, hindlimb steering, and center-of-gravity adjustment to enable robust performance in both environments. Through kinematic modeling, dynamic analysis, fluid-structure interaction simulations, and prototype testing, we demonstrate the feasibility and functionality of our bionic robot, contributing to advancements in adaptive robotic systems.
The development of bionic robots has evolved from manufacturing applications to areas like construction, rescue, military, and healthcare. However, most existing robots are limited by environmental adaptability. For example, many underwater robots cannot operate on land, complicating deployment and recovery. To overcome this, researchers have explored amphibious designs based on animal locomotion, such as snake-like蠕动 or leg-based systems. While these offer flexibility, they often compromise stability or efficiency. The sea turtle’s flapping propulsion—a hybrid of avian flight and fish swimming—provides a promising solution, offering high burst capability, maneuverability, and stability. Our goal is to harness these traits in a bionic robot capable of transitioning between water and land without external assistance. This work details our approach to designing such a bionic robot, emphasizing kinematic analysis, mechanical optimization, and experimental validation.
Sea turtles exhibit remarkable amphibious abilities, using their front flippers for propulsion in water and adapting them for crawling on land. Underwater, the flippers perform a flapping motion consisting of two phases: a power stroke, where the flipper is perpendicular to the direction of travel and pushes backward rapidly to generate thrust, and a recovery stroke, where the flipper is parallel to the direction and moves forward slowly to minimize drag. On land, the flippers fold inward, reducing the distance between the contact points and the body’s center of gravity, facilitating stable crawling. This dual-mode operation inspires our bionic robot’s core design. Additionally, sea turtles achieve turning through asymmetric flipper movements and buoyancy control by adjusting flipper angles. For simplicity, our bionic robot employs synchronized flipper motions for straight-line propulsion, with steering handled by a hindlimb mechanism and buoyancy regulated via a center-of-gravity adjustment system. This holistic approach ensures the bionic robot mimics key biological functions while maintaining mechanical reliability.
Our amphibious bionic robot comprises four main mechanisms: the flapping transmission mechanism, the deformable self-locking hydrofoil mechanism, the hindlimb mechanism, and the center-of-gravity adjustment mechanism. The overall design prioritizes compactness, durability, and energy efficiency, with a shell-like body housing all components. The flapping transmission mechanism converts rotary motor input into oscillatory flipper motions, simulating the sea turtle’s power and recovery strokes. The hydrofoil mechanism allows the flippers to transition between extended and folded states for water and land locomotion, respectively, using a self-locking spring-based system. The hindlimb mechanism enables turning by generating lateral impulses, while the center-of-gravity mechanism adjusts the robot’s pitch angle for diving and surfacing. Together, these systems empower the bionic robot to navigate complex environments autonomously. Below, we detail each mechanism’s design and function, supported by analytical models and simulations.
The flapping transmission mechanism is central to the bionic robot’s aquatic propulsion. It consists of a central planar four-bar linkage coupled with two SRRR spatial linkages. The planar four-bar linkage transforms the constant rotation of a dual-shaft DC geared motor into an elliptical motion with quick-return characteristics, optimizing the flipper’s velocity profile. The SRRR spatial linkages further convert this motion into a flapping action, ensuring that during the power stroke, the flipper is perpendicular to the direction of travel for maximum thrust, and during the recovery stroke, it is parallel to minimize drag. The linkage parameters are optimized to replicate the sea turtle’s flapping kinematics. We use double-layer linkages and multiple bearings to enhance strength and stability. The motor’s torque is transmitted symmetrically via gears to both sides, ensuring balanced forces and smooth operation. This design allows the bionic robot to achieve efficient underwater propulsion, a critical feature for amphibious functionality.
The deformable self-locking hydrofoil mechanism enables the bionic robot to adapt its flippers for terrestrial crawling. Each hydrofoil includes a shoulder plate, an outer plate, and a skeletal frame. In the aquatic extended state, the outer plate provides thrust during flapping. Upon landing, during the first flapping cycle, the flipper’s tip contacts the ground, triggering a transformation via a planar five-bar linkage and a self-locking spring. The mechanism passes a locking临界点, transitioning to a folded state where the shoulder plate contacts the ground for friction-based crawling. The self-locking spring ensures stability in both states, preventing unintended transitions. This automatic shape-shifting allows the bionic robot to switch environments without manual intervention, enhancing its autonomy as a versatile bionic robot.
The hindlimb mechanism facilitates steering by模仿 sea turtle tail movements. It comprises a parallelogram linkage and two RSSR spatial linkages. A servo motor drives the parallelogram, which transmits motion to the spatial linkages, causing the hindlimbs to flip laterally. At extreme positions, the hindlimbs reach near-vertical orientations, generating lateral impulses for turning. In neutral position, the hindlimbs remain horizontal, allowing straight-line motion. Compared to rudder systems, this design offers more biomimetic and responsive steering, crucial for navigating obstacles in water or on land. The mechanism is compact and low-power, aligning with the bionic robot’s overall energy constraints.
The center-of-gravity adjustment mechanism controls the bionic robot’s buoyancy and pitch angle. It includes a stepper motor,同步轮 and belt, linear导轨, and a movable counterweight. By shifting the counterweight along the导轨, the robot’s center of gravity relative to its buoyancy center changes, altering the pitch angle. This adjusts the hydrofoil’s angle of attack, enabling diving or surfacing when the robot is neutrally buoyant. We ensure the total weight matches the buoyant force for悬浮状态. This mechanism provides precise depth control, expanding the bionic robot’s operational range in aquatic environments. The use of a stepper motor allows fine-grained adjustments, enhancing stability during submerged missions.
To validate the flapping motion, we performed kinematic simulations using Inventor software. The planar four-bar linkage’s output follows an elliptical path, with velocity peaking at point A during the power stroke. The SRRR spatial linkages transform this into a flapping trajectory, amplifying velocity for effective thrust generation. The simulations confirm that the mechanism replicates the sea turtle’s two-phase stroke, with rapid backward motion and slow forward return. This analysis ensures the bionic robot can achieve efficient propulsion, a key milestone in developing amphibious capabilities. We further derived mathematical models to quantify flipper angles and motor requirements, as discussed below.
The flipper’s motion involves two degrees of freedom:公转 rotation around axis 1 and自转 rotation around axis 2. Let $\theta_3$ be the公转 angle (projection of axis 2 on the horizontal plane relative to the forward direction) and $\beta$ be the自转 angle (angle between the flipper plane and the plane defined by axis 2 and the forward direction). Both are functions of the input link angle $\alpha$. Using vector loop equations for the planar four-bar and SRRR linkages, we express $\theta_3$ and $\beta$ analytically. The governing equations are derived from geometric constraints:
For the planar four-bar linkage with link lengths $L_{01}$, $L_1$, $L_2$, $L_3$, $L_4$, $L_5$, and angles $\gamma_0$, $\theta_1$, $\theta_2$, $\gamma_1$, $\gamma_2$, we have:
$$L_1 \cos \alpha + L_2 \cos \theta_1 – L_{01} \cos \gamma_0 – L_3 \cos \theta_2 = 0$$
$$L_1 \sin \alpha + L_2 \sin \theta_1 – L_{01} \sin \gamma_0 – L_3 \sin \theta_2 = 0$$
Solving for $\theta_1$ and $\theta_2$, the point P coordinates are:
$$(0, y_p, z_p) = (0, L_1 \cos \alpha + L_4 \cos(\theta_1 + \gamma_1), L_1 \sin \alpha + L_4 \sin(\theta_1 + \gamma_1))$$
where $\gamma_1 = \arccos\left(\frac{L_2^2 + L_4^2 – L_5^2}{2L_2L_4}\right)$ and $\gamma_2 = \arccos\left(\frac{L_2^2 + L_5^2 – L_4^2}{2L_2L_5}\right)$.
For the SRRR linkage with lengths $L_6$, $L_7$, and angles $\gamma_3$, $\theta_4$, $\theta_5$, the equations are:
$$L_6 \cos \theta_5 \cos \theta_4 + L_7 \cos \gamma_3 \cos \theta_3 – L_{04} = 0$$
$$L_6 \cos \theta_5 \sin \theta_4 + L_7 \cos \gamma_3 \sin \theta_3 – L_{02} + y_p = 0$$
$$L_6 \sin \theta_5 + L_7 \sin \gamma_3 + L_{03} – z_p = 0$$
Solving these yields $\theta_3$, $\theta_4$, $\theta_5$. The自转 angle $\beta$ is then:
$$\beta = \pi – \arccos\left(\frac{(\mathbf{L}_6 \times \mathbf{L}_7) \cdot \mathbf{k}}{|\mathbf{L}_6 \times \mathbf{L}_7| |\mathbf{k}|}\right) – \gamma_4$$
where $\mathbf{k}$ is the unit vector along axis 2 and $\gamma_4$ is the flipper installation angle. Plotting $\theta_3$ and $\beta$ versus $\alpha$ reveals three phases: power stroke (negative $\theta_3$, $\beta$ near 90°), flip phase ($\theta_3 \approx 0$, $\beta$ changing), and recovery stroke (positive $\theta_3$, $\beta \approx 0°). This confirms the bionic robot’s ability to mimic biological flapping.
The motor torque required for flapping depends on hydrodynamic forces. We model the water reaction as a concentrated load at the hydrofoil’s centroid, with area $S$ and distances $L_8$ from axis 1 and $\Delta l$ from axis 2. The velocity component normal to the flipper plane is:
$$v = \frac{dx}{dt} = \frac{\Delta l \, d\beta_3 – L_8 \cos \gamma_3 \, d\theta_3 \sin \beta}{dt}$$
The thrust force is approximated as $F_0 = \rho S v^2$, where $\rho$ is water density. Through force analysis of the linkages, the motor torque $M$ is derived:
$$M = \left[ F_x \left( \sin \alpha + \frac{L_4 \sin(\theta_1 + \gamma_1) \sin(\theta_2 – \alpha)}{L_2 \sin(\theta_2 – \theta_1)} \right) + F_y \left( \cos \alpha + \frac{L_4 \cos(\theta_1 + \gamma_1) \sin(\theta_2 – \alpha)}{L_2 \sin(\theta_2 – \theta_1)} \right) \right] L_1 i$$
where $F_x$ and $F_y$ are constraint forces from the planar linkage, and $i$ is the gear ratio. For our parameters, the maximum torque is 0.022 N·m, well within the motor’s rated 3.92 N·m, ensuring reliable operation of the bionic robot.
Proper weight distribution is crucial for buoyancy control. We calculate fixed counterweight placement to achieve a 45° pitch angle when the movable counterweight is at its extreme position (250 mm travel). Let $p_w$ be the fixed counterweight mass, $(p_x, p_y)$ its coordinates relative to the buoyancy center, $a_w$ the total mass, $k_w$ the frame mass, $(k_x, k_y)$ the frame’s center, $(r_x, r_y)$ the movable counterweight’s center, and $d_x$ its travel distance. The关系 are:
$$p_w = \frac{r_x \cdot a_w + d_x \cdot a_w + a_w \cdot r_y – (r_y + r_x + d_x) \cdot k_w + k_y \cdot k_w + k_x \cdot k_w}{(r_y + r_x + d_x – p_y – p_x)}$$
The vertical distance between center of gravity and buoyancy center is:
$$a_y = \frac{p_w \cdot p_y + k_w \cdot k_y + r_y \cdot (a_w – p_w – k_w)}{a_w}$$
The movable counterweight mass is $r_w = a_w – p_w$. Using MATLAB, we optimize for stability ( $a_y > 15$ mm) and practical mass limits ( $r_w < 10$ kg). The feasible region yields an optimal fixed counterweight position at (-20 mm, -15 mm), ensuring the bionic robot remains stable during maneuvers.
To assess structural integrity, we conducted fluid-structure interaction analysis on the hydrofoil outer plate using ANSYS and Fluent. The aluminum plate was subjected to flapping at 2.09 rad/s angular velocity. The deformation and stress云图 show maximum deformation of 17.672 mm at the tip and maximum stress of 53.965 MPa at the root, below aluminum’s yield strength. However, the plate’s mass (1.405 kg) impacts agility. We explored two alternatives: an acrylic一体式 plate (0.600 kg) and a hybrid design with aluminum骨架 and acrylic skin (0.969 kg). The acrylic plate exhibited excessive stress (96.009 MPa), while the hybrid design reduced deformation (19.676 mm tip) and kept stresses within limits (42.239 MPa for acrylic, 238.54 MPa for aluminum). Thus, we adopted the hybrid方案 for the bionic robot, balancing weight and strength.
| Parameter | Value |
|---|---|
| Robot dimensions (without flippers) | 377 mm × 360 mm × 260 mm |
| Total mass | 36.3 kg |
| Max underwater speed | 0.5 m/s |
| Max terrestrial speed | 0.3 m/s |
| Max水下俯仰角 | 45° |
| Flipper flapping frequency | 2 Hz |
| Motor power rating | 24 V DC, 50 W |
We fabricated a prototype of the bionic robot to validate our design. Testing confirmed basic functionalities: underwater flapping propulsion, stable forward movement, turning via hindlimbs, and pitch control up to 45°. The hydrofoils successfully transitioned between extended and folded states, enabling crawling on land. However, we encountered issues: variable deformation of underwater flexible seals due to pressure differences, and ground contact resistance during terrestrial locomotion. To address seal deformation, we added limiters to prevent over-expansion. For ground resistance, we installed one-way hinges on the flipper tips, allowing propulsion during the power stroke while minimizing drag during recovery. These modifications improved the bionic robot’s performance, demonstrating practical adaptability.
The development of this amphibious bionic robot highlights the potential of biomimicry in creating versatile robotic systems. Our design integrates multiple mechanisms to replicate sea turtle locomotion, enabling operation in both aquatic and terrestrial environments. Kinematic and dynamic analyses ensure efficient flapping motion, while fluid-structure simulations guide material optimization. Prototype testing validates core functions and informs iterative improvements. Future work will focus on enhancing hydrodynamic efficiency through shape optimization, reducing weight for better agility, and developing reusable sealing solutions for easier maintenance. This bionic robot represents a step toward autonomous machines capable of navigating complex, dynamic environments, with applications in environmental monitoring, search and rescue, and exploration.
In conclusion, we have presented a comprehensive design and analysis of an amphibious bionic robot inspired by sea turtles. By leveraging flapping propulsion, deformable hydrofoils, and adaptive control mechanisms, this bionic robot achieves robust performance across water and land. The integration of analytical models, simulations, and experimental testing ensures reliability and functionality. As bionic robots continue to evolve, such biomimetic approaches will play a crucial role in overcoming environmental limitations, paving the way for more autonomous and adaptable robotic systems. We believe this work contributes valuable insights to the field of amphibious robotics and encourages further exploration of nature-inspired designs.