Swimming strategy of settling elongated micro-swimmers by reinforcement learning
- PDF / 1,328,418 Bytes
- 9 Pages / 595.276 x 793.701 pts Page_size
- 22 Downloads / 153 Views
gust 2020 Vol. 63 No. 8: 284711 https://doi.org/10.1007/s11433-019-1502-2
Swimming strategy of settling elongated micro-swimmers by reinforcement learning JingRan Qiu, WeiXi Huang, ChunXiao Xu, and LiHao Zhao
*
Applied Mechanics Laboratory (AML), Department of Engineering Mechanics, Tsinghua University, 100084 Beijing, China Received November 22, 2019; accepted December 30, 2019; published online April 3, 2020
Particular types of plankton in aquatic ecosystems can coordinate their motion depending on the local flow environment to reach regions conducive to their growth or reproduction. Investigating their swimming strategies with regard to the local environment is important to obtain in-depth understanding of their behavior in the aquatic environment. In the present research, to examine an impact of the shape and gravity on a swimming strategy, plankton is considered as settling swimming particles of ellipsoidal shape. The Q-learning approach is adopted to obtain swimming strategies for smart particles with a goal of efficiently moving upwards in a two-dimensional steady flow. Strategies obtained from reinforcement learning are compared to those of naive gyrotactic particles that are modeled considering the behavior of realistic plankton. It is found that the elongation of particles improves the performance of upward swimming by facilitating particles’ resistance to the perturbation of vortex. In the case when the settling velocity is included, the strategy obtained by reinforcement learning has similar performance to that of the naive gyrotactic one, and they both align swimmers in upward direction. The similarity between the strategy obtained from machine learning and the biological gyrotactic strategy indicates the relationship between the aspherical shape and settling effect of realistic plankton and their gyrotactic feature. swimming particles, ellipsoid, settling, reinforcement learning PACS number(s): 47.55.nb, 47.20.Ky, 47.11.Fg Citation:
J. R. Qiu, W. X. Huang, C. X. Xu, and L. H. Zhao, Swimming strategy of settling elongated micro-swimmers by reinforcement learning, Sci. ChinaPhys. Mech. Astron. 63, 284711 (2020), https://doi.org/10.1007/s11433-019-1502-2
1 Introduction Various plankton are ubiquitous in the marine environment. Many of them are able to swim and coordinate their motion by using special cellular structures, such as flagella [1,2]. The impact of the swimming motion on the distribution and deposition of plankton in the flow has been considered as an important question. Active plankton are often considered as swimming particles, which can rotate themselves toward a particular direction and swim along it accordingly [3,4].
Most of the previous studies have been focusing on a naive particle model, in which the particles are supposed to be motile, but are unable to sense or react with the ambient flow environment. Gyrotactic particles, for example, adjust their alignment by a gravitational torque induced by the bias of centers of mass and hydrodynamic force [3]. The motion and deposition of
Data Loading...
