Traditional underwater propulsion relies heavily on rigid propellers. While effective for high-speed transit, propellers are often noisy, inefficient at low speeds, and lack the agility required for complex environments. In contrast, aquatic organisms have evolved sophisticated locomotion strategies using flexible bodies and multiple coordinated fins.
Recent research focuses on replicating these biological advantages by moving beyond simple flapping mechanisms toward integrated systems that coordinate multiple control surfaces and exploit structural flexibility.
The Advantage of Multi-Fin Systems
Most early robotic fish focused on a single propulsor, typically a caudal (tail) fin. However, biological fish utilize a suite of fins—pectoral, pelvic, dorsal, and anal—to achieve high-maneuverability and stability. Yu et al. demonstrate that integrating multiple control surfaces allows for gaits that are impossible for single-fin robots, such as precise sideward motion and enhanced stability during complex maneuvers.
This coordination is often managed through Central Pattern Generators (CPGs), which mimic the neural networks in animals that produce rhythmic movements. By combining CPGs with visual feedback or advanced control frameworks like Deep Reinforcement Learning (DRL), researchers can achieve stable target-following and schooling behaviors. For instance, Wang et al. show that predictive reinforcement learning can reduce overshooting and improve the efficiency of these biomimetic systems.
Exploiting Structural Flexibility
A critical realization in this field is that a fin’s performance is not just a function of its movement, but of its material properties. Biological fins are not uniformly rigid; they possess specific distributions of stiffness that allow them to deform advantageously under fluid pressure.
Research into these "bionic bones" reveals that the ratio of stiffness across different axes of the fin is paramount:
- Spanwise vs. Chordwise Stiffness: Cao et al. found that while overall stiffness matters at low speeds, spanwise stiffness (along the length of the fin) becomes the dominant factor for thrust at higher speeds.
- Leading vs. Trailing Edges: By studying manta ray-inspired propulsion, Cao et al. determined that a stiff leading edge combined with a flexible trailing edge significantly improves hydrodynamic efficiency. This gradient allows the fin to "scoop" and push water more effectively than a rigid plate.
As Li et al. note in their review, replicating this inherent flexibility is one of the most significant challenges in modern soft robotic design, as it requires balancing mechanical simplicity with the complex wave propagation seen in nature.
Scaling and Environment Adaptation
The principles of bio-inspired propulsion extend across scales and environments. At the micro-scale, researchers look to flagellated organisms to solve the challenges of low-Reynolds-number environments, where water feels as viscous as honey. Shi et al. describe how mimicking bacterial flagella allows magnetic microrobots to achieve efficient helical propulsion for targeted drug delivery. Similarly, Du et al. utilize discrete elastic rod simulations to control soft flagellated robots navigating air-fluid interfaces.
Furthermore, the transition between environments has led to the development of amphibious robots. Cheng et al. introduced "FroBot," which uses a dual-swing-legs mechanism to transition from land movement to underwater undulation. This versatility suggests that the future of bio-inspired robotics lies not just in mimicking a single species, but in synthesizing various biological mechanisms—from the burrowing efficiency of moles (Tirtawardhana) to the vision-less navigation of bats (Pikalov et al.)—to create robust, autonomous systems.
Key Foundational Work
Design and Control of an Embedded Vision Guided Robotic Fish with Multiple Control Surfaces (2014) — contributed 4 concept elements to this direction.
Design, modeling and control of a novel amphibious robot with dual-swing-legs propulsion mechanism (2015) — contributed 3 concept elements to this direction.