In recent years, the field of robotics has seen significant advancements, particularly in the development of bionic robots that mimic biological organisms. Among these, multi-legged robots, such as hexapod bionic robots, have garnered attention due to their superior adaptability to unstructured terrains. As a researcher in this domain, I have focused on enhancing the mobility and stability of hexapod bionic robots through innovative gait design and simulation. This study presents the development of a hexapod bionic robot based on Raspberry Pi visual navigation, aiming to improve its autonomous crawling capabilities over rough terrain. The mechanical structure was designed using SolidWorks, and gait models were established to derive kinematic equations. Through simulation and experimental validation, we demonstrate the effectiveness of our approach in achieving stable and efficient locomotion for bionic robots.
The hexapod bionic robot is inspired by multi-legged insects, such as spiders and ants, which exhibit remarkable stability and agility. Bionic robots, in general, are categorized into humanoid, biomimetic, and bio-robots, with multi-legged designs excelling in non-structured environments due to their continuous walking ability. Our hexapod bionic robot consists of a main body, mechanical arms, six arc-shaped legs, and two hexagonal fuselages. The control system integrates 22 CDS5516 motors, a Raspberry Pi controller, a camera, and various sensors. The components are evenly distributed to ensure the robot’s center of gravity remains at the core, enhancing stability. The leg design features three links with lengths denoted as L1, L2, and L3, each comprising three joints and three limb segments. The use of a hexagonal fuselage, as opposed to a square one, was chosen to maximize stability by providing a larger support polygon, as shown in the comparison where the distance D2 > D1.
To quantify the mechanical parameters, Table 1 summarizes the components of our hexapod bionic robot. The materials and dimensions were selected to balance strength and weight, critical for efficient locomotion in bionic robots.
| Component | Material | Length (mm) | Mass (kg) |
|---|---|---|---|
| Hexagonal Fuselage | Aluminum Alloy | 400 | 0.80 |
| Link L1 | Aluminum Alloy | 5 | 0.02 |
| Link L2 | Aluminum Alloy | 100 | 0.42 |
| Link L3 | Aluminum Alloy | 130 | 0.56 |
| Anti-slip Foot Cover | ABS | 20 | 0.01 |
| Mechanical Gripper | Aluminum Alloy | 80 | 0.40 |
The CDS5516 motors used in our bionic robot provide high torque and precision, as detailed in Table 2. These motors enable smooth joint movements, essential for implementing complex gaits in bionic robots.
| Parameter | Value |
|---|---|
| Torque | 1.568 N·m |
| Output Speed | 0.16 s/60° |
| DC Voltage | 6.0–16 V |
| Position Resolution | 0.32° |
| Serial Units | 254 |
| Communication Baud Rate | 1 M |
| Servo Update Rate | 0.25 kHz |
Gait design is fundamental to the locomotion of hexapod bionic robots. We developed three primary straight-line gaits—tripod, quadruped, and pentapod—along with a turning gait. The tripod gait, the fastest and most efficient, involves alternating两组 of legs (1,4,5 and 2,3,6) in swing and support phases. The quadruped gait offers a balance between speed and stability, with up to six feet in contact during motion. The pentapod gait, though slowest, provides maximum stability and load-bearing capacity, ideal for rough terrain. For turning, we designed an in-place gait where legs rotate around the center point, enabling smooth directional changes. These gaits enhance the adaptability of bionic robots in diverse environments.
To model the locomotion, we established a D-H coordinate system for the leg joints, as illustrated in Figure 1. The D-H parameters are listed in Table 3, which serve as the basis for kinematic analysis. The forward kinematics equation derives the end-effector position from joint angles, while inverse kinematics computes joint angles from desired positions. This kinematic model is crucial for trajectory planning in bionic robots.
| Link n | Joint Angle θn | Twist Angle αn | Link Offset dn | Link Length an |
|---|---|---|---|---|
| 1 | θ1 | 90° | 0 | L1 |
| 2 | θ2 | 0° | 0 | L2 |
| 3 | θ3 | 0° | 0 | L3 |
The transformation matrix between links is given by:
$$ A_n = \begin{bmatrix} \cos\theta_n & -\sin\theta_n\cos\alpha_n & \sin\theta_n\sin\alpha_n & a_n\cos\theta_n \\ \sin\theta_n & \cos\theta_n\cos\alpha_n & -\cos\theta_n\sin\alpha_n & a_n\sin\theta_n \\ 0 & \sin\alpha_n & \cos\alpha_n & d_n \\ 0 & 0 & 0 & 1 \end{bmatrix} $$
For our bionic robot, substituting the D-H parameters yields:
$$ A_1 = \begin{bmatrix} \cos\theta_1 & 0 & \sin\theta_1 & L_1\cos\theta_1 \\ \sin\theta_1 & 0 & -\cos\theta_1 & L_1\sin\theta_1 \\ 0 & 1 & 0 & 0 \\ 0 & 0 & 0 & 1 \end{bmatrix} $$
$$ A_2 = \begin{bmatrix} \cos\theta_2 & -\sin\theta_2 & 0 & L_2\cos\theta_2 \\ \sin\theta_2 & \cos\theta_2 & 0 & L_2\sin\theta_2 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 1 \end{bmatrix} $$
$$ A_3 = \begin{bmatrix} \cos\theta_3 & -\sin\theta_3 & 0 & L_3\cos\theta_3 \\ \sin\theta_3 & \cos\theta_3 & 0 & L_3\sin\theta_3 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 1 \end{bmatrix} $$
The overall forward kinematics equation is:
$$ T_3 = A_1 A_2 A_3 = \begin{bmatrix} n_x & o_x & a_x & P_x \\ n_y & o_y & a_y & P_y \\ n_z & o_z & a_z & P_z \\ 0 & 0 & 0 & 1 \end{bmatrix} $$
where:
$$ P_x = \cos\theta_1[L_3\cos(\theta_2 + \theta_3) + L_2\cos\theta_2 + L_1] $$
$$ P_y = \sin\theta_1[L_3\cos(\theta_2 + \theta_3) + L_2\cos\theta_2 + L_1] $$
$$ P_z = L_3\sin(\theta_2 + \theta_3) + L_2\sin\theta_2 $$
For inverse kinematics, given the end-effector position (P_x, P_y, P_z), we solve for joint angles. From the equations, we derive:
$$ \theta_1 = \arctan\left(\frac{P_y}{P_x}\right) $$
$$ \theta_3 = \arccos\left(\frac{\cos^2\theta_1(P_z^2 – 2P_z + L_1^2) + P_x^2 – 2P_x L_1\cos\theta_1 – \cos^2\theta_1(L_3^2 + L_2^2)}{2L_2 L_3 \cos^2\theta_1}\right) $$
$$ \theta_2 = \arcsin\left(\frac{P_z}{\sqrt{L_2^2 + 2L_2 L_3 \cos\theta_3 + L_3^2}}\right) – \arctan\left(\frac{L_3 \sin\theta_3}{L_2 + L_3 \cos\theta_3}\right) $$
These equations enable precise control of the bionic robot’s leg movements during gait execution.
Gait planning involves defining trajectories for the swing and support phases. For the support phase, the foot-end moves along a straight line relative to the body, driven by ground friction. Assuming Z = 0, the trajectory is parallel to the Y-axis, with X = 120 mm based on our design. Substituting the link lengths (L1 = 5 mm, L2 = 100 mm, L3 = 130 mm) into the inverse kinematics equations, we compute joint angles as functions of stride length. Using MATLAB, we plotted these relationships, as shown in Figure 2, which illustrates the variation of joint angles with stride length for the support phase. This analysis ensures smooth motion in bionic robots.
For the swing phase, the trajectory is parabolic to facilitate obstacle clearance. The foot-end follows a path defined by quadratic functions in the Y-Z and Y-X planes. Let h_z and h_x be the vertex heights, and k_z and k_x the slopes. The equations are:
$$ k_z = -\frac{4h_z}{m^2}, \quad z = k_z y^2 + h_z $$
$$ k_x = \frac{4(L – h_x)}{m^2}, \quad x = k_x y^2 + h_x $$
where m is the stride length. For adaptive步长, we adjust the vertex heights based on stride changes Δm:
$$ \Delta h_z = k_z\left(\frac{m – \Delta m}{2}\right)^2 + h_z, \quad z = k_z y^2 + h_z – \Delta h_z $$
$$ \Delta h_x = k_x\left(\frac{m – \Delta m}{2}\right)^2 + h_x, \quad x = k_x y^2 + h_x – \Delta h_x $$
The resulting trajectory, plotted in Figure 3, shows a smooth curve for foot clearance. The joint angles during the swing phase are derived from inverse kinematics and plotted in Figure 4, demonstrating coordinated movements for the bionic robot.
Simulation验证 was conducted using ADAMS to assess the stability of straight-line and turning gaits. We set parameters such as a stride length of 300 mm, leg lift of 80 mm, simulation time of 10 s, and 200 steps. The driving functions were based on trigonometric equations to minimize冲击. For turning, we varied the stride ratios between left and right sides. Figure 5 compares the body centroid displacement trajectories for right stride multipliers of 1.2 and 1.5. The results indicate that a larger right stride multiplier reduces the turning radius, enhancing agility in bionic robots. Additionally, reducing the leg width to 0.5 times the original size slightly decreased speed but maintained stability, as shown in Figure 6. These simulations confirm the robustness of our gait designs for bionic robots.
To further analyze performance, we compiled simulation data in Table 4, summarizing key metrics like stability margin and speed for each gait. This comparative analysis aids in selecting appropriate gaits for specific terrains in bionic robot applications.
| Gait Type | Max Support Legs | Speed (mm/s) | Stability Margin | Applications |
|---|---|---|---|---|
| Tripod | 3 | High | Low | Fast traversal on flat terrain |
| Quadruped | 6 | Medium | Medium | Moderate loads and uneven ground |
| Pentapod | 6 | Low | High | Heavy loads and rough terrain |
| Turning | Varies | Adjustable | High | Directional changes in confined spaces |
Experimental validation involved building a physical prototype of the hexapod bionic robot, as depicted in Figure 7. The robot has 18 degrees of freedom, with joint angles controlled by gait algorithms. We tested the motor performance by executing a 300° rotation program and recording execution counts versus speed. The relationship, plotted in Figure 8, shows that higher speeds reduce execution counts and precision, highlighting the need for optimized control in bionic robots. The实物测试 confirmed that our gait designs enable stable crawling, with the robot successfully navigating obstacles and maintaining balance. This practical verification underscores the effectiveness of our approach for real-world bionic robot deployments.
In terms of broader implications, bionic robots like ours have potential in search-and-rescue, planetary exploration, and agricultural automation. The integration of Raspberry Pi for visual navigation allows for autonomous decision-making, a key advancement in bionic robot technology. Future work could focus on dynamic gait adaptation using machine learning or incorporating more sensors for enhanced environmental perception. Moreover, optimizing the mechanical structure for weight reduction could improve energy efficiency in bionic robots.
From a kinematic perspective, we expanded the analysis to include velocity and acceleration profiles. Using the Jacobian matrix derived from the forward kinematics, we computed the end-effector velocity as:
$$ \mathbf{v} = \mathbf{J}(\mathbf{\theta}) \dot{\mathbf{\theta}} $$
where $\mathbf{J}$ is the Jacobian, and $\dot{\mathbf{\theta}}$ is the joint velocity vector. For our bionic robot, the Jacobian for a single leg is:
$$ \mathbf{J} = \begin{bmatrix} -L_3 s_1 c_{23} – L_2 s_1 c_2 – L_1 s_1 & -L_3 c_1 s_{23} – L_2 c_1 s_2 & -L_3 c_1 s_{23} \\ L_3 c_1 c_{23} + L_2 c_1 c_2 + L_1 c_1 & -L_3 s_1 s_{23} – L_2 s_1 s_2 & -L_3 s_1 s_{23} \\ 0 & L_3 c_{23} + L_2 c_2 & L_3 c_{23} \end{bmatrix} $$
where $s_i = \sin\theta_i$, $c_i = \cos\theta_i$, and $s_{23} = \sin(\theta_2 + \theta_3)$, $c_{23} = \cos(\theta_2 + \theta_3)$. This allows for dynamic control and obstacle avoidance in bionic robots.
Additionally, we explored energy consumption metrics for the bionic robot. The power required by each motor can be estimated as:
$$ P = \tau \dot{\theta} $$
where $\tau$ is the torque and $\dot{\theta}$ is the angular velocity. Table 5 lists the average power consumption for different gaits, derived from simulation data. This information is vital for optimizing battery life in autonomous bionic robots.
| Gait Type | Average Power per Motor (W) | Total Power (W) | Energy Efficiency |
|---|---|---|---|
| Tripod | 2.1 | 37.8 | High |
| Quadruped | 2.5 | 45.0 | Medium |
| Pentapod | 3.0 | 54.0 | Low |
To enhance the bionic robot’s adaptability, we implemented a polynomial interpolation method for gait transition. The joint angles during phase changes are fitted using fifth-degree polynomials:
$$ \theta(t) = a_0 + a_1 t + a_2 t^2 + a_3 t^3 + a_4 t^4 + a_5 t^5 $$
where coefficients are determined based on boundary conditions (position, velocity, acceleration). This ensures smooth transitions between swing and support phases, reducing mechanical stress on the bionic robot.
In conclusion, this study presents a comprehensive framework for autonomous crawling gait design and simulation of a hexapod bionic robot. We developed multiple gaits, derived kinematic models, and validated performance through simulations and experiments. The results demonstrate that our bionic robot achieves stable and efficient locomotion, with applications in challenging environments. Future directions include integrating AI for adaptive gait control and miniaturizing components for swarm robotics. The continuous advancement of bionic robot technology holds promise for revolutionizing fields like disaster response and exploration, and our work contributes to this growing body of knowledge. By emphasizing the关键词 “bionic robot” throughout, we highlight the biomimetic essence of our design, which mimics biological principles to solve engineering challenges. Ultimately, this research underscores the potential of bionic robots as versatile platforms for autonomous operations in unstructured terrains.