Kundan Panta, Hankun Deng, Zhiyu Zhang, Daning Huang, Azar Panah, Bo Cheng
{"title":"Touchless underwater wall-distance sensing via active proprioception of a robotic flapper.","authors":"Kundan Panta, Hankun Deng, Zhiyu Zhang, Daning Huang, Azar Panah, Bo Cheng","doi":"10.1088/1748-3190/ad2114","DOIUrl":null,"url":null,"abstract":"<p><p>In this work, we explored a bioinspired method for underwater object sensing based on active proprioception. We investigated whether the fluid flows generated by a robotic flapper, while interacting with an underwater wall, can encode the distance information between the wall and the flapper, and how to decode this information using the proprioception within the flapper. Such touchless wall-distance sensing is enabled by the active motion of a flapping plate, which injects self-generated flow to the fluid environment, thus representing a form of active sensing. Specifically, we trained a long short-term memory (LSTM) neural network to predict the wall distance based on the force and torque measured at the base of the flapping plate. In addition, we varied the Rossby number (<i>Ro</i>, or the aspect ratio of the plate) and the dimensionless flapping amplitude (A∗) to investigate how the rotational effects and unsteadiness of self-generated flow respectively affect the accuracy of the wall-distance prediction. Our results show that the median prediction error is within 5% of the plate length for all the wall-distances investigated (up to 40 cm or approximately 2-3 plate lengths depending on the<i>Ro</i>); therefore, confirming that the self-generated flow can enable underwater perception. In addition, we show that stronger rotational effects at lower<i>Ro</i>lead to higher prediction accuracy, while flow unsteadiness (A∗) only has moderate effects. Lastly, analysis based on SHapley Additive exPlanations (SHAP) indicate that temporal features that are most prominent at stroke reversals likely promotes the wall-distance prediction.</p>","PeriodicalId":55377,"journal":{"name":"Bioinspiration & Biomimetics","volume":null,"pages":null},"PeriodicalIF":3.1000,"publicationDate":"2024-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Bioinspiration & Biomimetics","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.1088/1748-3190/ad2114","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0
Abstract
In this work, we explored a bioinspired method for underwater object sensing based on active proprioception. We investigated whether the fluid flows generated by a robotic flapper, while interacting with an underwater wall, can encode the distance information between the wall and the flapper, and how to decode this information using the proprioception within the flapper. Such touchless wall-distance sensing is enabled by the active motion of a flapping plate, which injects self-generated flow to the fluid environment, thus representing a form of active sensing. Specifically, we trained a long short-term memory (LSTM) neural network to predict the wall distance based on the force and torque measured at the base of the flapping plate. In addition, we varied the Rossby number (Ro, or the aspect ratio of the plate) and the dimensionless flapping amplitude (A∗) to investigate how the rotational effects and unsteadiness of self-generated flow respectively affect the accuracy of the wall-distance prediction. Our results show that the median prediction error is within 5% of the plate length for all the wall-distances investigated (up to 40 cm or approximately 2-3 plate lengths depending on theRo); therefore, confirming that the self-generated flow can enable underwater perception. In addition, we show that stronger rotational effects at lowerRolead to higher prediction accuracy, while flow unsteadiness (A∗) only has moderate effects. Lastly, analysis based on SHapley Additive exPlanations (SHAP) indicate that temporal features that are most prominent at stroke reversals likely promotes the wall-distance prediction.
期刊介绍:
Bioinspiration & Biomimetics publishes research involving the study and distillation of principles and functions found in biological systems that have been developed through evolution, and application of this knowledge to produce novel and exciting basic technologies and new approaches to solving scientific problems. It provides a forum for interdisciplinary research which acts as a pipeline, facilitating the two-way flow of ideas and understanding between the extensive bodies of knowledge of the different disciplines. It has two principal aims: to draw on biology to enrich engineering and to draw from engineering to enrich biology.
The journal aims to include input from across all intersecting areas of both fields. In biology, this would include work in all fields from physiology to ecology, with either zoological or botanical focus. In engineering, this would include both design and practical application of biomimetic or bioinspired devices and systems. Typical areas of interest include:
Systems, designs and structure
Communication and navigation
Cooperative behaviour
Self-organizing biological systems
Self-healing and self-assembly
Aerial locomotion and aerospace applications of biomimetics
Biomorphic surface and subsurface systems
Marine dynamics: swimming and underwater dynamics
Applications of novel materials
Biomechanics; including movement, locomotion, fluidics
Cellular behaviour
Sensors and senses
Biomimetic or bioinformed approaches to geological exploration.