Abstract Biological microswimmers can coordinate their motions to exploit their fluid environment—and each other—to achieve global advantages in their locomotory performance. These cooperative locomotion require delicate adjustments of both individual swimming gaits and spatial arrangements of the swimmers. Here we probe the emergence of such cooperative behaviors among artificial microswimmers endowed with artificial intelligence. We present the first use of a deep reinforcement learning approach to empower the cooperative locomotion of a pair of reconfigurable microswimmers. The AI-advised cooperative policy comprises two stages: an approach stage where the swimmers get in close proximity to fully exploit hydrodynamic interactions, followed a synchronization stage where the swimmers synchronize their locomotory gaits to maximize their overall net propulsion. The synchronized motions allow the swimmer pair to move together coherently with an enhanced locomotion performance unattainable by a single swimmer alone. Our work constitutes a first step toward uncovering intriguing cooperative behaviors of smart artificial microswimmers, demonstrating the vast potential of reinforcement learning towards intelligent autonomous manipulations of multiple microswimmers for their future biomedical and environmental applications.
more »
« less
Gait switching and targeted navigation of microswimmers via deep reinforcement learning
Abstract Swimming microorganisms switch between locomotory gaits to enable complex navigation strategies such as run-and-tumble to explore their environments and search for specific targets. This ability of targeted navigation via adaptive gait-switching is particularly desirable for the development of smart artificial microswimmers that can perform complex biomedical tasks such as targeted drug delivery and microsurgery in an autonomous manner. Here we use a deep reinforcement learning approach to enable a model microswimmer to self-learn effective locomotory gaits for translation, rotation and combined motions. The Artificial Intelligence (AI) powered swimmer can switch between various locomotory gaits adaptively to navigate towards target locations. The multimodal navigation strategy is reminiscent of gait-switching behaviors adopted by swimming microorganisms. We show that the strategy advised by AI is robust to flow perturbations and versatile in enabling the swimmer to perform complex tasks such as path tracing without being explicitly programmed. Taken together, our results demonstrate the vast potential of these AI-powered swimmers for applications in unpredictable, complex fluid environments.
more »
« less
- NSF-PAR ID:
- 10348369
- Date Published:
- Journal Name:
- Communications Physics
- Volume:
- 5
- Issue:
- 1
- ISSN:
- 2399-3650
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Microorganisms are commonly found swimming in complex biological fluids such as mucus and these fluids respond elastically to deformation. These viscoelastic fluids have been previously shown to affect the swimming kinematics of these microorganisms in non-trivial ways depending on the rheology of the fluid, the particular swimming gait and the structural properties of the immersed body. In this report we put forth a previously unmentioned mechanism by which swimming organisms can experience a speed increase in a viscoelastic fluid. Using numerical simulations and asymptotic theory we find that significant swirling flow around a microscopic swimmer couples with the elasticity of the fluid to generate a marked increase in the swimming speed. We show that the speed enhancement is related to the introduction of mixed flow behind the swimmer and the presence of hoop stresses along its body. Furthermore, this effect persists when varying the fluid rheology and when considering different swimming gaits. This, combined with the generality of the phenomenon (i.e. the coupling of vortical flow with fluid elasticity near a microscopic swimmer), leads us to believe that this method of speed enhancement could be present for a wide range of microorganisms moving through complex fluids.more » « less
-
The use of machine learning techniques in the development of microscopic swimmers has drawn considerable attention in recent years. In particular, reinforcement learning has been shown useful in enabling swimmers to learn effective propulsion strategies through its interactions with the surroundings. In this work, we apply a reinforcement learning approach to identify swimming gaits of a multi-link model swimmer. The swimmer consists of multiple rigid links connected serially with hinges, which can rotate freely to change the relative angles between neighboring links. Purcell [“Life at low Reynolds number,” Am. J. Phys. 45, 3 (1977)] demonstrated how the particular case of a three-link swimmer (now known as Purcell's swimmer) can perform a prescribed sequence of hinge rotation to generate self-propulsion in the absence of inertia. Here, without relying on any prior knowledge of low-Reynolds-number locomotion, we first demonstrate the use of reinforcement learning in identifying the classical swimming gaits of Purcell's swimmer for case of three links. We next examine the new swimming gaits acquired by the learning process as the number of links increases. We also consider the scenarios when only a single hinge is allowed to rotate at a time and when simultaneous rotation of multiple hinges is allowed. We contrast the difference in the locomotory gaits learned by the swimmers in these scenarios and discuss their propulsion performance. Taken together, our results demonstrate how a simple reinforcement learning technique can be applied to identify both classical and new swimming gaits at low Reynolds numbers.more » « less
-
Microorganisms can efficiently navigate in anisotropic complex fluids, but the precise swimming mechanisms remain largely unexplored. Their dynamics are determined by the interplay between multiple effects, including the fluid's orientation order, swimmer's undulatory gait, and the finite length. Here we extend the numerical study of the two-dimensional undulatory motions of a flexible swimmer in lyotropic liquid-crystalline polymers (LCPs) by Lin et al. (2021) to the scenarios of arbitrary swimming directions with respect to the nematic director. The swimmer is modeled as a nearly inextensible yet flexible fiber with imposed traveling-wave like actuation. We investigate the orientation-dependent swimming behaviors in nematic LCPs for an infinite long sheet (i.e., Taylor's swimming sheet model) and finite-length swimmers. We demonstrate that the swimmer must be sufficiently stiff to produce undulatory deformations to gain net motions. Moreover, a motile finite-length swimmer can reorient itself to swim parallel with the nematic director, due to a net body torque arising from the asymmetric distribution of the polymer force along the body.more » « less
-
Soft robotics holds tremendous potential for various applications, especially in unstructured environments such as search and rescue operations. However, the lack of autonomy and teleoperability, limited capabilities, absence of gait diversity and real-time control, and onboard sensors to sense the surroundings are some of the common issues with soft-limbed robots. To overcome these limitations, we propose a spatially symmetric, topologically-stable, soft-limbed tetrahedral robot that can perform multiple locomotion gaits. We introduce a kinematic model, derive locomotion trajectories for different gaits, and design a teleoperation mechanism to enable real-time human-robot collaboration. We use the kinematic model to map teleoperation inputs and ensure smooth transitions between gaits. Additionally, we leverage the passive compliance and natural stability of the robot for toppling and obstacle navigation. Through experimental tests, we demonstrate the robot's ability to tackle various locomotion challenges, adapt to different situations, and navigate obstructed environments via teleoperation.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1038/s42005-022-00935-x
