[Paper Review] Detecting exotic wakes with hydrodynamic sensors

Mengying Wang, Maziar S. Hemati|arXiv (Cornell University)|Nov 28, 2017

Biomimetic flight and propulsion mechanisms5 references3 citations

TL;DR

This paper proposes a general wake detection protocol that classifies exotic hydrodynamic wakes—beyond standard von Kármán-type wakes—using hydrodynamic signals measured at a single point on a fish-like body. By modeling wake dynamics via an ideal-flow framework and extracting time-series signatures, the method achieves over 95% accuracy in distinguishing between dynamically distinct wake regimes, demonstrating viability for multi-model sensing in bioinspired robotics.

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 Karman-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.

Motivation & Objective

• To address the lack of effective wake sensing strategies for exotic wake types that deviate from standard von Kármán-type dynamics.
• To develop a general-purpose detection framework capable of distinguishing between wake regimes based solely on hydrodynamic signals.
• To demonstrate the feasibility of classifying complex, non-2S wake dynamics using a feature-based classification library.
• To evaluate the robustness and accuracy of the detection protocol across diverse wake parameter regimes and algorithmic settings.

Proposed method

• Formulated an ideal-flow model to simulate hydrodynamic responses of a fish-like body to various wake types.
• Collected time-series hydrodynamic signals at a single surface point on the body to extract wake signatures.
• Extracted discriminative features from the time-series data to represent distinct wake dynamics.
• Constructed a wake classification library using feature vectors from a wide range of known wake regimes.
• Applied a k-nearest-neighbor algorithm to classify unknown wakes by comparing their signatures to the library entries.
• Evaluated performance across varying algorithm parameters to assess robustness and accuracy.

Experimental results

Research questions

• RQ1 Can hydrodynamic signals measured at a single point on a body reliably distinguish between exotic wake types with different underlying dynamics?
• RQ2 How effective is a feature-based classification library in identifying unknown wake regimes from hydrodynamic signatures?
• RQ3 What is the accuracy of the detection protocol across diverse wake parameter values and algorithmic configurations?
• RQ4 What are the limitations of the method in distinguishing between wake regimes near phase-space separatrices?
• RQ5 How do signal characteristics and noise levels affect the performance of the wake classification framework?

Key findings

• The wake detection protocol achieved an accuracy rate exceeding 95% across the majority of tested algorithm parameter values.
• Distinct wake dynamics impart unique hydrodynamic signatures on a fish-like body, observable in time-series measurements at a single point.
• Feature vectors effectively captured qualitative differences between dynamically distinct wake regimes, enabling reliable classification.
• The method struggled to distinguish between wake regimes evolving near phase-space separatrices, where feature vectors were more similar.
• Performance degradation was observed in regimes with mixed characteristics, such as those exhibiting traits of both orbiting and exchanging dynamics.
• The study highlights the need for richer classification libraries and multi-modal, distributed sensing to address real-world challenges like sensor noise and variable body-wake alignment.