In the rapidly evolving field of robotics, the study of bionic technology has deeply permeated and subtly influenced our daily lives. My research focuses on multi-legged robots, which simulate the crawling motions of animals through multi-joint structures. In practical applications, robots with more than four legs are collectively termed multi-legged robots. The investigation into multi-legged bionic robots remains a hot topic globally. In challenging terrains, wheeled and tracked robots face significant limitations. To enhance operational capabilities in complex or adverse conditions, the development and study of multi-legged robotic devices hold immense importance. Through my research on multi-legged bionic robots and detailed observations of crocodile locomotion, I have designed a quadruped structure that mimics crocodile leg movements. This structure integrates limbs and motors to form a multi-legged bionic device based on an Arduino control core. The bionic crocodile features a flexible tail with multiple servo joints, with the main body comprising four legs, a torso, a tail, and a head. The legs are critical for motion, each consisting of six primary structures. Through shafts, the front and hind limbs are coordinately controlled. Upon receiving commands, the servo controller drives the robot’s leg movements via transmission shafts, enabling forward motion and steering by adjusting motor speeds. This amphibious bionic crocodile can operate both on land and underwater.
The background and significance of this study stem from the limitations of traditional wheeled and tracked robots. As demands for robots with high stability and strong adaptability grow in production and daily life, robots capable of navigating恶劣 and complex terrains have gained prominence. With ongoing research, the motion stability and precision of multi-legged robots have improved substantially. Bionic robots offer distinct advantages: they better adapt to varied terrains and environments, utilize multi-joint structures for coordinated multi-degree-of-freedom movements, enhancing stability and flexibility. By employing spherical contacts with the ground, bionic robots reduce contact area, ensuring smoother operation on uneven surfaces and boosting terrain adaptability. This unique bionic structure allows these robots to simulate crocodile motion effectively.
In the overall system design, I employ an Arduino-based control core. The ARM microcontroller on the main hardware board executes 32-bit instructions, enabling rapid servo control. The system achieves stable motion through a triangular gait. Based on GF set theory, each leg is designed with three main joints. A servo and transmission shaft coordinate the front limbs, with a similar setup for the hind limbs. During locomotion, at least two legs maintain ground contact as supports. This design enhances motion stability, prevents interference between legs, and maximizes leg movement space. Under the premise that the angle α between the forward direction line and the line segment at the robot’s base center remains within a small range, increasing the distance between the robot’s center of gravity and its projection on the base improves stability. On complex surfaces, intelligent gait adjustments can be made—for instance, computing optimal gaits via computer algorithms for stable walking. In modeling, I use SOLIDWORKS software. Initially, I create an overall model to simulate the device’s motion. Then, I design individual parts, export drawings for processing, and finally assemble components to complete the structure.
For leg structure analysis, I adopt a 9-bar linkage with one degree of freedom, as shown in the figure. This simple, reliable, and durable design physiologically mimics skeletal forms such as the ankle, knee, and hip joints. By simulating real animal motion structures, the servo drives the main shaft rotation, which, via bevel gears, rotates the transmission shaft to generate leg movement. Electric drive is employed for its cleanliness, low noise, precise control, and strong anti-interference capabilities. Recent advancements in motors, such as high-torque, high-speed variants, have significantly propelled research in bionic robots. Leg optimization is conducted using MATLAB tools, performing finite element analysis on each unit and node to derive optimal structural configurations.
All joint structures in this bionic robot are controlled by motors, with linkages made of aluminum alloy to enhance stability and flexibility. The overall design enables the multi-legged robot to effectively simulate reptile gaits and motions. This structure boosts robot stability, reduces collisions between legs during movement, and expands leg motion space. Typically, connecting a servo’s control line to a microcontroller pin allows PWM control. However, for complex multi-legged devices with multi-joint structures, I enhance the PWM method to control multiple servos simultaneously via multi-channel PWM signals. Since motor controllers lack autonomous decision-making, pre-set motion instructions are required; the servo controller acts as a slave receiving commands. The control core uses an STC89C53 microcontroller to send commands, while the motor control board generates control signals to coordinate movements. Computer-written serial communication programs transmit control instructions to the microcontroller, enabling specific gait actions. Based on trajectory and position detection, the Arduino main board sends speed commands to motors for automatic adjustments.
The application of finite element analysis in this mechanism is extensive for structural optimization. First, I discretize the structure into independent small units, analyzing nodal displacements and forces for each. By exploring force-displacement relationships and combining units, I formulate equations for overall nodal displacements and forces. Solving these equations yields unknown nodal displacements, which are then substituted back into units for numerical solutions. In this bionic robot, I use SOLIDWORKS to analyze displacement, strain, and stress in leg components. For example, consider a linkage connected to a shaft in the leg structure. The motor used is a ZLSZ42 integrated closed-loop stepper motor with a holding torque of 0.72 N·m. Through transmission mechanisms, the force on the linkage is calculated as a 60 N·m moment and 50 N normal force. After discretizing the linkage in SOLIDWORKS, I fix the surface connected to the shaft and apply forces to obtain stress, displacement, and strain data. The discretized model and results are summarized below.
To illustrate the optimization process, I present key equations and tables. The motion of the bionic robot can be described using kinematic equations. For a leg with three joints, the position of the foot relative to the body is given by:
$$ \mathbf{p} = f(\theta_1, \theta_2, \theta_3) $$
where $\theta_1$, $\theta_2$, and $\theta_3$ are joint angles. The Jacobian matrix $\mathbf{J}$ relates joint velocities to foot velocity:
$$ \dot{\mathbf{p}} = \mathbf{J} \dot{\mathbf{\theta}} $$
For stability analysis, the center of gravity projection is crucial. The stability margin $S$ can be expressed as:
$$ S = \min(d_i) $$
where $d_i$ are distances from the center of gravity projection to the edges of the support polygon. In finite element analysis, the stress-strain relationship follows Hooke’s law for linear elasticity:
$$ \sigma = \mathbf{E} \epsilon $$
where $\sigma$ is stress, $\mathbf{E}$ is the elasticity matrix, and $\epsilon$ is strain. The nodal force-displacement equation is:
$$ \mathbf{F} = \mathbf{K} \mathbf{u} $$
with $\mathbf{F}$ as the force vector, $\mathbf{K}$ as the stiffness matrix, and $\mathbf{u}$ as the displacement vector.
Below is a table summarizing the leg joint parameters for the bionic robot:
| Joint | Type | Range of Motion (degrees) | Servo Model |
|---|---|---|---|
| Hip | Revolute | -30 to 30 | ZLSZ42 |
| Knee | Revolute | 0 to 60 | ZLSZ42 |
| Ankle | Revolute | -20 to 20 | ZLSZ42 |
Another table details the material properties used in finite element analysis:
| Component | Material | Young’s Modulus (GPa) | Density (kg/m³) |
|---|---|---|---|
| Linkage | Aluminum Alloy | 70 | 2700 |
| Shaft | Steel | 200 | 7850 |
The finite element analysis results for the linkage show maximum stress, displacement, and strain values. For instance, under the applied loads, the maximum von Mises stress $\sigma_{max}$ is computed as:
$$ \sigma_{max} = 45.6 \, \text{MPa} $$
which is below the yield strength of aluminum alloy (typically 200 MPa), indicating safety. The maximum displacement $u_{max}$ is:
$$ u_{max} = 0.12 \, \text{mm} $$
and the maximum strain $\epsilon_{max}$ is:
$$ \epsilon_{max} = 0.00065 $$
These values ensure the bionic robot’s components withstand operational forces without failure. By iteratively analyzing each leg component, I optimize the design for durability and performance. This process highlights the importance of finite element methods in developing robust bionic robots.
In terms of control algorithms, I implement gait patterns using periodic functions. For a triangular gait, the foot trajectory for each leg can be modeled as:
$$ x(t) = A \sin(\omega t + \phi) $$
$$ z(t) = B \cos(\omega t + \phi) $$
where $A$ and $B$ are amplitudes, $\omega$ is angular frequency, and $\phi$ is phase offset. The coordination between legs ensures continuous support. The phase differences $\Delta \phi$ for a quadruped bionic robot are:
$$ \Delta \phi_{12} = \pi, \quad \Delta \phi_{23} = \pi/2, \quad \Delta \phi_{34} = \pi $$
This results in stable locomotion. Additionally, I use PWM duty cycle $D$ to control servo angles:
$$ \theta = k \cdot D $$
with $k$ as a proportionality constant. For multi-servo control, the Arduino generates PWM signals via timers, with duty cycles adjusted based on gait calculations.
To enhance adaptability, I incorporate sensor feedback. For example, inertial measurement units (IMUs) provide orientation data, allowing real-time gait adjustments. The control law for speed adjustment is:
$$ \omega_{motor} = K_p e + K_i \int e \, dt $$
where $e$ is the error between desired and actual velocity, and $K_p$, $K_i$ are PID gains. This enables the bionic robot to navigate uneven terrain smoothly.
The amphibious capability of this bionic robot involves sealing components and optimizing propulsion. In water, the tail’s flexible joints generate thrust through oscillatory motions. The thrust force $F_t$ can be approximated as:
$$ F_t = C \rho A v^2 $$
where $C$ is a coefficient, $\rho$ is water density, $A$ is tail area, and $v$ is tail velocity. This dual-mode functionality expands the bionic robot’s application scope, such as in underwater exploration or disaster response.
In conclusion, my research on the bionic reptile robot demonstrates significant advancements in multi-legged bionic robotics. Through innovative design, simulation, and analysis, I have developed a device that mimics crocodile locomotion with high stability and adaptability. The use of finite element methods ensures structural integrity, while control algorithms enable efficient gait generation. This bionic robot exemplifies the potential of bionic technology in creating versatile robotic systems for complex environments. Future work may focus on autonomy, machine learning-based gait optimization, and enhanced sensor integration to further improve the bionic robot’s capabilities.
Throughout this study, the term “bionic robot” has been central, emphasizing the biomimetic approach. The bionic robot’s design leverages natural principles to achieve robust performance. As bionic robots evolve, they promise to revolutionize fields like search and rescue, environmental monitoring, and industrial automation. My ongoing efforts aim to refine this bionic robot, exploring new materials and control strategies to push the boundaries of what bionic robots can achieve.