Detecting exotic wakes with hydrodynamic sensors

Abstract

Wake sensing for bioinspired robotic swimmers has been the focus of much investigation owing to its relevance to locomotion control, especially in the context of schooling and target following. Many successful wake sensing strategies have been devised based on models of von Kármán-type wakes; however, such wake sensing technologies are invalid in the context of exotic wake types that commonly arise in swimming locomotion. Indeed, exotic wakes can exhibit markedly different dynamics, and so must be modeled and sensed accordingly. Here, we propose a general wake detection protocol for distinguishing between wake types from measured hydrodynamic signals alone. An ideal-flow model is formulated and used to demonstrate the general wake detection framework in a proof-of-concept study. We show that wakes with different underlying dynamics impart distinct signatures on a fish-like body, which can be observed in time-series measurements at a single location on the body surface. These hydrodynamic wake signatures are used to construct a wake classification library that is then used to classify unknown wakes from hydrodynamic signal measurements. The wake detection protocol is found to have an accuracy rate of over 95% in the majority of performance studies conducted here. Thus, exotic wake detection is shown to be viable, which suggests that such technologies have the potential to become key enablers of multi-model sensing and locomotion control strategies in the future.

Machine Learning for Fluid Mechanics

“…Recently, Colvert et al (2018) investigated the classification of wake topology (e.g., 2S, 2P+2S, 2P+4S) behind a pitching airfoil from local vorticity measurements using neural networks; extensions have compared performance for various types of sensors (Alsalman et al 2018). In Wang & Hemati (2017) the k nearest neighbors (KNN) algorithm was used to detect exotic wakes. Similarly, neural networks have been combined with dynamical systems models to detect flow disturbances and estimate their parameters (Hou et al 2019).…”

Classifying Exotic Wakes with a Flow Speed Sensor

“…After constructing the library, we use a supervised learning strategy to classify an unknown wake by comparing its hydrodynamic signals with the entries in the library. The synthesis of this protocol is following our previous work, 4 which is summarized graphically in Fig. 1.…”

“…In our previous work, we showed that this feature vector provided a convenient and relatively robust summary of the time-series signature imparted by various idealized wakes in our flow model. 4 In that study, we used the fast Fourier transform to extract the frequency content of the time-domain signal from each of the r wake realizations used to build-up the library {V 1 , V 2 , . .…”

“…9,10 Using this approach, a classification accuracy rate of 95% can be attained for a wide range of algorithm parameter values. 4 The description of the wake classification protocol above is generally applicable, but was not demonstrated in the context of datastreams. The flexibility to handle datastreams is an important consideration to make wake detection and classification more relevant for real-time applications.…”

“…In a recent study, we formulated a general framework for wake detection and classification. 4 The protocol was established as a supervised learning problem, in which a wake classification library was first constructed, then used to classify unknown wake types from time-series measurements of velocity at a single point on a body. The framework was demonstrated on an ideal flow model in a simple proof-of-concept setting, and was found to yield an accuracy rate of 95% over a wide range of algorithmic parameters.…”