In the field of underwater robotics, reducing operational resistance is a critical challenge that directly impacts energy efficiency and overall performance. As robot technology advances, the integration of mechanical arms into underwater inspection systems has become essential for tasks such as maintenance, sampling, and data collection. However, these additions can significantly increase drag, leading to higher energy consumption and reduced operational range. This article presents a comprehensive study on optimizing the installation position and angle of mechanical arms on an underwater inspection robot to minimize drag, leveraging biomimetic design principles and computational fluid dynamics (CFD) simulations. By focusing on robot technology enhancements, we aim to improve the efficiency and functionality of autonomous underwater vehicles (AUVs) in complex marine environments.
The development of underwater robot technology has evolved to include various types, such as remotely operated vehicles (ROVs) and autonomous underwater vehicles (AUVs), with AUVs being particularly valued for their flexibility and intelligence in inspection tasks. A key aspect of modern robot technology is the integration of mechanical arms, which enable precise manipulation and interaction with underwater structures. However, the addition of these arms often introduces additional drag, which can compromise the robot’s hydrodynamic performance. To address this, we employed a biomimetic approach, using the Zebra Pufferfish as a model due to its stable hovering capabilities and low-drag morphology. Through three-dimensional reconstruction and CFD analysis, we identified optimal installation parameters for the mechanical arms, ensuring minimal resistance during巡航 and operation. This research underscores the importance of robot technology in achieving sustainable and efficient underwater operations.
Our methodology began with the reconstruction of the Zebra Pufferfish model using structure-from-motion (SFM) techniques and PhotoScan software. This process involved capturing multiple images from different angles, processing them to generate point cloud data, and refining the model to a manageable mesh for simulation. The resulting biomimetic model retained the key hydrodynamic features of the fish, providing a foundation for the robot’s main body. Subsequently, we analyzed the workspace for the mechanical arms, considering factors such as visibility, propulsion interference, and operational range. Statistical analysis of the workspace volume helped identify the optimal installation points on the robot’s body. Finally, we conducted CFD simulations to evaluate the drag effects at various arm angles, leading to the identification of configurations that minimize resistance. This integrated approach highlights the role of advanced robot technology in solving real-world engineering challenges.
The three-dimensional reconstruction process involved several steps to ensure accuracy and usability. We collected 186 images of the Zebra Pufferfish from horizontal, 30-degree, and 60-degree angles, accounting for lighting variations. Using PhotoScan, we applied incremental SFM to compute camera positions and generate a dense point cloud. After filtering irrelevant data, we exported the mesh model and simplified it in Blender to reduce face count while preserving critical features. The final model had 170 faces, making it suitable for CFD simulations. The table below summarizes the key parameters from the reconstruction process, demonstrating the efficiency of this robot technology in capturing biological characteristics for engineering applications.
| Parameter | Value |
|---|---|
| Number of Images | 186 |
| Point Cloud Data Points | 17,059 |
| Key Point Limit | 600,000 |
| Total Mesh Faces | 18,029 |
| Valid Mesh Faces | 9,786 |
| Simplified Faces for CFD | 170 |
For the mechanical arm selection, we chose the Dongqi S5 model due to its compatibility with underwater robot technology, including a length of 600 mm, diameter of 50 mm, and capability to operate at depths up to 300 meters. This arm’s standardization and lightweight design make it ideal for integration into our biomimetic robot. To determine the optimal installation position, we divided the robot’s front section into 10 slices, each with 24 points, and performed Boolean intersection operations to calculate the workspace volume for each point. The results indicated that point 7 in group 10 offered the maximum workspace, as shown in the statistical analysis below. This method exemplifies how robot technology can be used to optimize component placement for enhanced functionality.
| Group | Point | Workspace Volume (Relative Units) |
|---|---|---|
| 1 | 1-10 | 0.45 – 0.78 |
| 2 | 1-10 | 0.50 – 0.82 |
| 3 | 1-10 | 0.55 – 0.85 |
| 4 | 1-10 | 0.60 – 0.88 |
| 5 | 1-10 | 0.65 – 0.90 |
| 6 | 1-10 | 0.70 – 0.92 |
| 7 | 1-10 | 0.75 – 0.94 |
| 8 | 1-10 | 0.80 – 0.96 |
| 9 | 1-10 | 0.85 – 0.98 |
| 10 | 7 | 1.00 (Maximum) |
With the installation position established, we focused on determining the optimal angle for the mechanical arms to minimize drag during巡航. We used CFD simulations in Fluent, setting up a flow domain of 4 m × 3 m × 3 m with water as the fluid medium. The RNG k-ε turbulence model was employed due to its accuracy in handling complex flows, with the governing equations given by:
$$ \frac{\partial (\rho k)}{\partial t} + \frac{\partial (\rho k u_i)}{\partial x_i} = \frac{\partial}{\partial x_j} \left( \alpha_k u_{\text{eff}} \frac{\partial k}{\partial x_j} \right) + G_k + G_b – \rho \varepsilon – Y_M + S_k $$
$$ \frac{\partial (\rho \varepsilon)}{\partial t} + \frac{\partial (\rho \varepsilon u_i)}{\partial x_i} = \frac{\partial}{\partial x_j} \left( \alpha_\varepsilon u_{\text{eff}} \frac{\partial \varepsilon}{\partial x_j} \right) + G_{1\varepsilon} \frac{\varepsilon}{k} (G_k + G_{3\varepsilon} G_b) – G_{2\varepsilon} \rho \frac{\varepsilon^2}{k} + S_\varepsilon – R_\varepsilon $$
Here, \( u_{\text{eff}} \) represents the effective velocity, \( k \) is the turbulent kinetic energy, \( \varepsilon \) is the dissipation rate, and other terms account for various production and dissipation effects. The simulations were conducted for 30 different arm angles from 6° to 180°, with the resistance force calculated using the standard drag equation:
$$ F = \frac{1}{2} C \rho S V^2 $$
where \( C \) is the drag coefficient, \( \rho \) is the fluid density, \( S \) is the reference area, and \( V \) is the velocity. The results, summarized in the table below, show that angles between 132° and 156° yielded the lowest drag, with the minimum resistance of 0.0303 N occurring at 150°. This finding is crucial for advancing robot technology, as it provides a quantitative basis for optimizing mechanical arm configurations.
| Group | Angle (°) | Resistance (N) |
|---|---|---|
| 1 | 6 | 0.0431 |
| 2 | 12 | 0.0473 |
| 3 | 18 | 0.0511 |
| 4 | 24 | 0.0532 |
| 5 | 30 | 0.0545 |
| 6 | 36 | 0.0561 |
| 7 | 42 | 0.0564 |
| 8 | 48 | 0.0574 |
| 9 | 54 | 0.0593 |
| 10 | 60 | 0.0574 |
| 11 | 66 | 0.0564 |
| 12 | 72 | 0.0549 |
| 13 | 78 | 0.0531 |
| 14 | 84 | 0.0505 |
| 15 | 90 | 0.0484 |
| 16 | 96 | 0.0457 |
| 17 | 102 | 0.0431 |
| 18 | 108 | 0.0430 |
| 19 | 114 | 0.0402 |
| 20 | 120 | 0.0365 |
| 21 | 126 | 0.0350 |
| 22 | 132 | 0.0322 |
| 23 | 138 | 0.0312 |
| 24 | 144 | 0.0304 |
| 25 | 150 | 0.0303 |
| 26 | 156 | 0.0306 |
| 27 | 162 | 0.0330 |
| 28 | 168 | 0.0353 |
| 29 | 174 | 0.0376 |
| 30 | 180 | 0.0400 |
The CFD simulations revealed that drag was significantly higher at angles between 12° and 96°, with an average resistance above 0.045 N, while the optimal range of 132° to 156° had resistances below 0.033 N. This demonstrates how robot technology can leverage simulation tools to achieve performance gains. The velocity and pressure distribution云图 from the simulations further illustrated that the mechanical arm’s angle affects flow separation and wake formation, with minimal disruption at 150°. These insights are vital for designing energy-efficient underwater robots, as reducing drag can extend operational time and range, key metrics in robot technology applications.
Based on these findings, we developed a comprehensive design for the underwater inspection robot. The main body, inspired by the Zebra Pufferfish, features a tubular frame with环形支撑架 for modular component installation. Critical modules, such as cameras, sensors, and control systems, are housed in the front and middle sections, while power and buoyancy control units are placed on the sides. The mechanical arms are symmetrically installed at the optimal position (point 7, group 10) and set to 150° during巡航 to minimize drag. In operation, the arms can deploy forward for tasks while the robot hovers using仿生 fins and thrusters. This design exemplifies the integration of robot technology with biomimetic principles to create adaptive and efficient systems.
In conclusion, this study highlights the importance of optimizing mechanical arm installation in underwater robot technology to reduce drag and enhance efficiency. Through biomimetic modeling, workspace analysis, and CFD simulations, we identified the optimal position and angle for the arms, resulting in a significant reduction in resistance. The proposed design incorporates these findings, enabling the robot to operate effectively in both巡航 and task modes. Future work will explore the drag effects during active manipulation, further advancing robot technology for underwater applications. Overall, this research contributes to the ongoing development of intelligent and sustainable robotic systems, underscoring the transformative potential of robot technology in marine exploration and industry.