美国北卡罗来纳州立大学的研究人员开发出一种新型软体水下机器人，其外形与运动方式受到海洋生物羽冠海百合的启发。与许多需要多个电机和复杂控制系统才能实现立体运动的水下机器人不同，这款设备仅使用两个气动驱动输入，就能完成上升、下降、前后移动、悬停以及绕自身轴线旋转等动作。
这项研究的核心并不只是模仿海洋生物的外形，而是尝试将生物运动中蕴含的机械原理应用于机器人设计。羽冠海百合属于海百合类棘皮动物，具有多条柔软、羽毛状的腕肢，能够通过协调不同腕肢的运动，在水中改变方向或保持位置。研究团队希望借助类似的结构设计，解决传统水下机器人在三维机动能力方面的局限。
几年前，同一研究团队曾开发过一种以蝠鲼为灵感的水下机器人。该设备能够快速游动，但在上下移动、原地转向以及复杂空间中的姿态调整方面存在不足。研究人员因此转向研究羽冠海百合，希望在不显著增加机械复杂度的情况下，让机器人获得更加灵活的运动能力。
新型机器人由一个中央主体和四条柔性翼状结构组成。中央部分集成了两个独立的气动驱动通道，四条弹性翼肢则负责与水相互作用，产生推进力和旋转力矩。通过改变两个驱动通道的工作方式，机器人可以在三种不同的运动模式之间切换。
第一种模式被研究人员称为“水母模式”。当两个驱动器同时工作时，机器人四条翼肢会迅速向下弯曲，推动水流并使自身向上运动。驱动停止后，翼肢凭借自身的弹性恢复到原来的形状，机器人随之改变运动状态并向下移动。通过调整动作频率，机器人既可以持续升降，也可以减缓运动，在水中保持近似悬停的状态。
第二种模式是“鱼类模式”。此时，研究人员只启动其中一个驱动通道，使机器人两侧的结构产生不对称运动。某一侧翼肢的摆动类似鱼类利用尾鳍推进的方式，能够推动机器人向前或向后移动。通过控制驱动输入，还可以改变推进动作，使机器人适应不同的移动方向。
第三种模式则是“转子模式”。研究人员通过交替控制两个驱动通道，使翼肢依次产生不同的运动，从而带动机器人绕自身轴线旋转。这种能力让机器人可以在水中调整朝向，为后续进行三维导航和定向操作提供了基础。
这三种模式结合起来，使机器人能够在上下、前后和旋转等方向上进行运动。研究团队指出，通常要实现类似的运动范围，传统设计可能需要至少六个驱动器，而这款机器人只需要两个。其关键在于一种被称为“机械智能”的设计理念。
所谓机械智能，是指通过结构本身的物理特性实现部分控制功能，而不是把所有动作都交给独立电机和复杂电子系统完成。在这款机器人中，柔性翼肢能够在受力弯曲后迅速恢复原状，结构中储存和释放的弹性能量帮助机器人产生快速、有规律的运动。研究人员利用这种特性，将有限的驱动输入转化为多种不同的运动形式。
这种设计还与软体机器人的发展方向密切相关。传统水下机器人通常依靠刚性结构、多个推进器和复杂的姿态控制系统来完成运动，而软体机器人则更多利用柔性材料的变形能力，与周围环境进行相互作用。对于需要在复杂水下环境中活动的设备而言，减少机械部件数量不仅有望简化结构，也可能降低控制和维护方面的负担。
在实验中，研究团队还分析了机器人运动时周围水流的变化。研究人员利用粒子图像测速技术观察水流运动，并结合计算流体力学模拟，研究不同运动模式下形成的涡流结构。这些分析有助于解释机器人如何通过翼肢摆动产生推进力，以及旋转动作如何形成相应的力矩。
实验结果显示，该机器人最高游动速度达到每秒1.64个自身身长，最高旋转速度为每秒90度。研究人员还报告了其运输成本指标的最低值为17.6，该指标用于衡量机器人移动所需的能量消耗。相关结果说明，利用简单结构实现多模式运动具有可行性，但这些实验数据并不意味着它已经达到实际海洋作业所需的全部性能要求。
研究团队展示了该平台的一些潜在用途，包括搭载水下摄像头探索环境、搬运或提升水下物体，以及让多个机器人协同工作。通过增加摄像头、夹持器或其他辅助设备，未来的机器人或许能够用于水下环境监测、海洋污染调查和水下物品回收等任务。
研究人员还设想，可以让多个机器人共同执行任务。例如，单台机器人负责观察和定位目标，其他机器人则协助搬运或收集物体。不过，这些应用目前仍属于潜在发展方向，机器人在真实海洋环境中的长期运行能力、续航时间、通信方式和复杂水流中的稳定性，还需要进一步验证。
目前，这款机器人仍依赖气动驱动输入。研究团队表示，后续工作将包括开发完全无线的版本，并与其他领域的研究人员合作，探索更具体的应用场景。无线化将有助于减少设备与外部控制系统之间的连接限制，但也会带来供能、控制和水下通信方面的新挑战。
这项研究成果发表于开放获取期刊《Science Advances》，论文题为《最少驱动的仿羽冠海百合软体游泳机器人实现多模式三维机动》。研究表明，机器人不一定要通过不断增加电机、传感器和控制系统来获得更强的机动能力。通过合理设计柔性结构，并充分利用材料本身的弹性和运动特性，也能够让相对简单的机械系统实现复杂动作。
对于水下机器人领域而言，这种思路可能具有更广泛的意义。未来，研究人员或许能够借鉴更多海洋生物的运动机制，开发结构更简单、运动方式更多样的软体机器人，使其在环境监测、水下探索和复杂空间作业等任务中发挥作用。
本文由 AI宝库自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
Researchers at the North Carolina State University in the United States have developed a new kind of soft submersible robot whose appearance and motion are inspired by the sea life plume. Unlike many underwater robots that require multiple engines and complex control systems in order to achieve a stereodynamic movement, this equipment can only use two aerodynamic-driven inputs to complete uplifting, falling, moving back and forth, suspension and rotation around its own axis.
The core of this study is not just to imitate the shape of marine life, but to try to apply the mechanical principles inherent in biological motion to robotic design. Sea lilies, which belong to the sea lilies and have many soft, feathered wrists, are able to change direction or maintain position in water by coordinating the movement of different wrists. The research team hopes to use similar structural designs to address the limitations of traditional underwater robots in terms of three-dimensional mobility.
A few years ago, the same research team developed an underwater robot inspired by thallium. The equipment is able to move quickly, but there are deficiencies in moving up and down, moving in situ and posturing in complex space. As a result, researchers have turned to research on the Sea lilies, hoping to give robots more flexible mobility without significantly increasing their mechanical complexity.
The new robot consists of a central subject and four flexible wing structures. The central part is grouped into two separate aerodynamic driveways, while the four flexi-wing limbs are responsible for interacting with water and generating propulsion and rotational rectangles. By changing the working methods of the two drivers, robots can switch between three different modes of motion.
The first model was referred to by researchers as the “meat model”. When the two drives work at the same time, the robots' four-wing limbs bend down quickly, push the water and move themselves up. When the drive stops, the wing limbs recover to their original shape with their own flexibility, and the robot changes his or her state of motion and moves downwards. By adjusting the frequency of the actions, robots can both keep up and down and slow their movement, maintaining near-suspension in water.
The second model is “fish patterns”. At this point, only one of the drivers is activated by researchers, creating asymmetrical movements in the structures on both sides of the robot. The swinging of a wing limb is similar to that of a fish using tail-fine propulsion and can drive robots forward or backward. By controlling the drive input, you can also change the move and adapt the robot to different movement directions.
The third model is a "rotational mode". Researchers rotate robots around their own axes by alternately controlling two drive lanes, resulting in different movements of wing limbs. This capability allows robots to adjust direction in water and provides a basis for subsequent 3D navigation and orientation operations.
These three models combine to enable robots to move up and down, back and forth, and rotate. The research team noted that traditional design may require at least six drives to achieve a similar scope of movement, and that this robot only needs two. The key is a design concept called “mechanical intelligence”.
Mechanical intelligence refers to partial control functions achieved through the physical properties of the structure itself, rather than leaving all actions to independent electric and complex electronic systems. In this robot, flexible wing limbs can be quickly restored to the state after the bend of the force, and the flexible energy stored and released in the structure helps the robot to generate rapid and regular motion. Using this feature, researchers have converted limited drivers into a variety of different forms of movement.
This design is also closely related to the direction of soft robots. Traditional underwater robots usually rely on rigid structures, multiple thrusters and complex attitude control systems to carry out their movements, while flexible robots are more able to interact with the surrounding environment using the deformation capacity of flexible materials. For equipment requiring activities in a complex underwater environment, reducing the number of mechanical components is expected not only to simplify the structure but also to reduce the control and maintenance burden.
In the experiment, the team also analysed changes in the water flow around robotics. Researchers use particle-image speed measurement techniques to observe current movements and to study vortex structures under different motion patterns in conjunction with computational fluid mechanic simulations. These analyses help to explain how robots generate propulsion through wing limb swings and how rotational actions form the corresponding force rectangles.
The results of the experiment showed that the robot had a maximum swim rate of 1.64 per second and a maximum rotation rate of 90 degrees per second. Researchers also reported a minimum value of 17.6 for their transport cost indicator, which measures the energy consumption required for robotic mobility. The results demonstrate the feasibility of using simple structures to achieve a multi-modal movement, but these experimental data do not mean that it has met all the performance requirements required for operational marine operations.
The research team demonstrated a number of potential uses of the platform, including carrying underwater cameras to explore the environment, moving or lifting underwater objects, and working with multiple robots. By adding cameras, handlers or other assistive devices, future robots may be able to be used in underwater environmental monitoring, marine pollution investigations and underwater item recovery tasks.
Researchers also envisage the possibility of multiple robots working together. For example, single robots observe and position targets, while other robots assist in moving or collecting objects. However, these applications are still a potential development, and the long-term operational capacity of robots in the real marine environment, the duration of their voyages, the means of communication and the stability of complex currents need further validation.
Currently, the robot still relies on aerodynamicly driven input. The team indicated that follow-up would include the development of a fully wireless version and the exploration of more specific applications in collaboration with researchers in other areas. Wirelessization will help to reduce the limitations of connectivity between equipment and external control systems, but it will also pose new challenges in energy supply, control and underwater communications.
The results of the study were published in the open-access journal Science Advances, a paper entitled " A multi-dimensional multi-dimensional motor for the least-drive mimic corona moss swimming robot ". Studies have shown that robots do not have to acquire greater mobility by increasing the number of electrics, sensors and control systems. Complex operations can also be made possible by the rational design of flexible structures and the full use of the material ' s own elasticity and motion characteristics.
This approach may have wider significance for underwater robotics. In the future, researchers may be able to draw on the motor mechanisms of more marine life and develop more simple-structured, mobile flexible robots to play a role in environmental monitoring, underwater exploration and complex space operations.
Auto-fetched and translated by AI Gem Box. All rights belong to the original author.