Pub Date : 2022-09-19DOI: 10.1109/AUV53081.2022.9965889
Milind Fernandes, S. R. Sahoo, Mangal Kothari
Thrusters are a vital part of any autonomous under-water or surface vehicle. However, due to their high costs, they are often out of reach of researchers and students in developing countries. While many designs are available using off-the-shelf drone motors, they are either not open source or do not provide performance details. This paper presents an open-source, 3D printed, low-cost, compact thruster designed using open-source software tools. We present our design and testing approach and the performance data gathered from experiments that are useful for modeling the thruster. The approach presented here can be applied to any off-the-shelf brushless motor to design a thruster if the one used in this paper is unavailable or does not meet specific performance criteria.
{"title":"An Open Source, Low Cost, 3D Printed Thruster For Autonomous Underwater and Surface Vehicles","authors":"Milind Fernandes, S. R. Sahoo, Mangal Kothari","doi":"10.1109/AUV53081.2022.9965889","DOIUrl":"https://doi.org/10.1109/AUV53081.2022.9965889","url":null,"abstract":"Thrusters are a vital part of any autonomous under-water or surface vehicle. However, due to their high costs, they are often out of reach of researchers and students in developing countries. While many designs are available using off-the-shelf drone motors, they are either not open source or do not provide performance details. This paper presents an open-source, 3D printed, low-cost, compact thruster designed using open-source software tools. We present our design and testing approach and the performance data gathered from experiments that are useful for modeling the thruster. The approach presented here can be applied to any off-the-shelf brushless motor to design a thruster if the one used in this paper is unavailable or does not meet specific performance criteria.","PeriodicalId":148195,"journal":{"name":"2022 IEEE/OES Autonomous Underwater Vehicles Symposium (AUV)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115006551","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-19DOI: 10.1109/AUV53081.2022.9965886
Mingxi Zhou, Jianguang Shi
Seabed mapping is a common application for marine robots, and it is often framed as a coverage path planning problem in robotics. During a robot-based survey, the coverage of perceptual sensors (e.g., cameras, LIDARS and sonars) changes, especially in underwater environments. Therefore, online path planning is needed to accommodate the sensing changes in order to achieve the desired coverage ratio. In this paper, we present a sensing confidence model and a uncertainty-driven sampling-based online coverage path planner (SO-CPP) to assist in-situ robot planning for seabed mapping and other survey-type applications. Different from conventional lawnmower pattern, the SO-CPP will pick random points based on a probability map that is updated based on in-situ sonar measurements using a sensing confidence model. The SO-CPP then constructs a graph by connecting adjacent nodes with edge costs determined using a multi-variable cost function. Finally, the SO-CPP will select the best route and generate the desired waypoint list using a multi-variable objective function. The SO-CPP has been evaluated in a simulation environment with an actual bathymetric map, a 6-DOF AUV dynamic model and a ray-tracing sonar model. We have performed Monte Carlo simulations with a variety of environmental settings to validate that the SO-CPP is applicable to a convex workspace, a non-convex workspace, and unknown occupied workspace. So-CPP is found outperform regular lawnmower pattern survey by reducing the resulting traveling distance by upto 20%. Besides that, we observed that the prior knowledge about the obstacles in the environment has minor effects on the overall traveling distance. In the paper, limitation and real-world implementation are also discussed along with our plan in the future.
{"title":"An Uncertainty-driven Sampling-based Online Coverage Path Planner for Seabed Mapping using Marine Robots","authors":"Mingxi Zhou, Jianguang Shi","doi":"10.1109/AUV53081.2022.9965886","DOIUrl":"https://doi.org/10.1109/AUV53081.2022.9965886","url":null,"abstract":"Seabed mapping is a common application for marine robots, and it is often framed as a coverage path planning problem in robotics. During a robot-based survey, the coverage of perceptual sensors (e.g., cameras, LIDARS and sonars) changes, especially in underwater environments. Therefore, online path planning is needed to accommodate the sensing changes in order to achieve the desired coverage ratio. In this paper, we present a sensing confidence model and a uncertainty-driven sampling-based online coverage path planner (SO-CPP) to assist in-situ robot planning for seabed mapping and other survey-type applications. Different from conventional lawnmower pattern, the SO-CPP will pick random points based on a probability map that is updated based on in-situ sonar measurements using a sensing confidence model. The SO-CPP then constructs a graph by connecting adjacent nodes with edge costs determined using a multi-variable cost function. Finally, the SO-CPP will select the best route and generate the desired waypoint list using a multi-variable objective function. The SO-CPP has been evaluated in a simulation environment with an actual bathymetric map, a 6-DOF AUV dynamic model and a ray-tracing sonar model. We have performed Monte Carlo simulations with a variety of environmental settings to validate that the SO-CPP is applicable to a convex workspace, a non-convex workspace, and unknown occupied workspace. So-CPP is found outperform regular lawnmower pattern survey by reducing the resulting traveling distance by upto 20%. Besides that, we observed that the prior knowledge about the obstacles in the environment has minor effects on the overall traveling distance. In the paper, limitation and real-world implementation are also discussed along with our plan in the future.","PeriodicalId":148195,"journal":{"name":"2022 IEEE/OES Autonomous Underwater Vehicles Symposium (AUV)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117061382","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-19DOI: 10.1109/AUV53081.2022.9965895
Alexandre Cardaillac, M. Ludvigsen
Underwater images are often degraded due to backscatter, light attenuation and light artifacts. One important aspect of it is marine snow, which are particles of varying shape and size. Computer vision technologies can be strongly affected by them and may therefore provide incorrect and biased results. In robotic applications, there is limited computational power for online processing. A method for real time marine snow detection is proposed in this paper based on a multi-step process of spatial-temporal data. The RGB colored images are converted to the YCbCr color space before they are decomposed to isolate the high frequency information using a guided filter for a first selection of candidates. Convolution with an uniform kernel is then applied for further analysis of the candidates. The method is demonstrated in two use cases, underwater feature detection and image enhancement.
{"title":"Marine Snow Detection for Real Time Feature Detection","authors":"Alexandre Cardaillac, M. Ludvigsen","doi":"10.1109/AUV53081.2022.9965895","DOIUrl":"https://doi.org/10.1109/AUV53081.2022.9965895","url":null,"abstract":"Underwater images are often degraded due to backscatter, light attenuation and light artifacts. One important aspect of it is marine snow, which are particles of varying shape and size. Computer vision technologies can be strongly affected by them and may therefore provide incorrect and biased results. In robotic applications, there is limited computational power for online processing. A method for real time marine snow detection is proposed in this paper based on a multi-step process of spatial-temporal data. The RGB colored images are converted to the YCbCr color space before they are decomposed to isolate the high frequency information using a guided filter for a first selection of candidates. Convolution with an uniform kernel is then applied for further analysis of the candidates. The method is demonstrated in two use cases, underwater feature detection and image enhancement.","PeriodicalId":148195,"journal":{"name":"2022 IEEE/OES Autonomous Underwater Vehicles Symposium (AUV)","volume":"130 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120922263","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-19DOI: 10.1109/AUV53081.2022.9965846
António J. Oliveira, B. Ferreira, R. Diamant, N. Cruz
In-situ calibration of marine sensors requires close-range positioning. In turn, localization relative to a given object of interest is necessary. This paper deals with the detection of a vertical cable hanging from a marine observatory implemented by means of a moored buoy. An algorithm composed of sequential image filtering, segmentation and template matching is proposed. Two approaches for generating the cable’s acoustic image template are introduced. The performance of the approaches, obtained by comparison with ground-truth measurements, are illustrated over challenging cluttered acoustic images collected in a test tank. The results indicate a performance better than 74% of the best candidate to match the actual cable.
{"title":"Sonar-based Cable Detection for in-situ Calibration of Marine Sensors","authors":"António J. Oliveira, B. Ferreira, R. Diamant, N. Cruz","doi":"10.1109/AUV53081.2022.9965846","DOIUrl":"https://doi.org/10.1109/AUV53081.2022.9965846","url":null,"abstract":"In-situ calibration of marine sensors requires close-range positioning. In turn, localization relative to a given object of interest is necessary. This paper deals with the detection of a vertical cable hanging from a marine observatory implemented by means of a moored buoy. An algorithm composed of sequential image filtering, segmentation and template matching is proposed. Two approaches for generating the cable’s acoustic image template are introduced. The performance of the approaches, obtained by comparison with ground-truth measurements, are illustrated over challenging cluttered acoustic images collected in a test tank. The results indicate a performance better than 74% of the best candidate to match the actual cable.","PeriodicalId":148195,"journal":{"name":"2022 IEEE/OES Autonomous Underwater Vehicles Symposium (AUV)","volume":"123 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127048667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-19DOI: 10.1109/AUV53081.2022.9965894
Alberto Consensi, Matthew Kingsland, Nick Linton, Leon Bowring, D. Roper, Richard Austin-Berry, Stewart Fairbairn, Alexis Johnson, Richard Morrison, Konrad Ciaramella, Daniel Matterson, M. Pebody, Val Williams, F. Fanelli, D. Fenucci, Achille Martin, Eoin Ó hÓbáin, Shivan Ramdhanie, Rashiid Sherif, Ashley Morris, A. Phillips
Autonomous Underwater Vehicles (AUVs) are proving to be a key component in the global observing system, with their ability to provide unique data sets particularly at abyssal depths or under ice. Autosub5 is the latest in a line of large work class AUVs developed by the National Oceanography Centre specifically tailored for oceanographic science applications. This paper describes the work currently being undertaken to transition the vehicle from an engineering prototype through to a science ready platform. The 18 months process saw the AUV assembled in early 2021 and then undertake a series of trials and incremental payload integrations through to a science rehearsal trial planned for summer 2022.
{"title":"Autosub5: Preparing for Science","authors":"Alberto Consensi, Matthew Kingsland, Nick Linton, Leon Bowring, D. Roper, Richard Austin-Berry, Stewart Fairbairn, Alexis Johnson, Richard Morrison, Konrad Ciaramella, Daniel Matterson, M. Pebody, Val Williams, F. Fanelli, D. Fenucci, Achille Martin, Eoin Ó hÓbáin, Shivan Ramdhanie, Rashiid Sherif, Ashley Morris, A. Phillips","doi":"10.1109/AUV53081.2022.9965894","DOIUrl":"https://doi.org/10.1109/AUV53081.2022.9965894","url":null,"abstract":"Autonomous Underwater Vehicles (AUVs) are proving to be a key component in the global observing system, with their ability to provide unique data sets particularly at abyssal depths or under ice. Autosub5 is the latest in a line of large work class AUVs developed by the National Oceanography Centre specifically tailored for oceanographic science applications. This paper describes the work currently being undertaken to transition the vehicle from an engineering prototype through to a science ready platform. The 18 months process saw the AUV assembled in early 2021 and then undertake a series of trials and incremental payload integrations through to a science rehearsal trial planned for summer 2022.","PeriodicalId":148195,"journal":{"name":"2022 IEEE/OES Autonomous Underwater Vehicles Symposium (AUV)","volume":"131 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134380198","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-19DOI: 10.1109/AUV53081.2022.9965795
P. Kampmann, C. Gaudig, Franka Nauert, M. Fritsche, T. Johannink
AUVs have been employed in underwater surveys for several years. These kinds of missions were mostly based on pre-defined waypoints in safe distances from obstacles and the seabed to generate maps of the environment. Current developments in the industry take the next step, long range AUVs are being designed by several manufacturers. The mission types range from survey missions to autonomous on-the-fly mission adaptations based on events and observations. These mission types require more advanced autonomy which should also be reflected in the software architecture of the AUV. An approach to tackle this is presented here.
{"title":"A software architecture for resilient long term autonomous missions of AUVs","authors":"P. Kampmann, C. Gaudig, Franka Nauert, M. Fritsche, T. Johannink","doi":"10.1109/AUV53081.2022.9965795","DOIUrl":"https://doi.org/10.1109/AUV53081.2022.9965795","url":null,"abstract":"AUVs have been employed in underwater surveys for several years. These kinds of missions were mostly based on pre-defined waypoints in safe distances from obstacles and the seabed to generate maps of the environment. Current developments in the industry take the next step, long range AUVs are being designed by several manufacturers. The mission types range from survey missions to autonomous on-the-fly mission adaptations based on events and observations. These mission types require more advanced autonomy which should also be reflected in the software architecture of the AUV. An approach to tackle this is presented here.","PeriodicalId":148195,"journal":{"name":"2022 IEEE/OES Autonomous Underwater Vehicles Symposium (AUV)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132423650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-19DOI: 10.1109/AUV53081.2022.9965905
Alessandro Bucci, Leonardo Zacchini, A. Ridolfi
Orientation estimation is a fundamental aspect of navigation and motion control of Autonomous Underwater Vehicles (AUVs). This concept is especially true when position sensors are unavailable and, consequently, navigation and control rely on dead reckoning strategies; in this case, orientation estimation is used in conjunction with speed measurements to update position estimation. When unknown magnetic disturbances are present, the magnetometers of the Inertial Measurement Unit (IMU) are unusable and do not provide an accurate initialization of the vehicle heading angle. This issue can be faced by applying a generalization of the Extended Kalman Filter in which the system state and measurements evolve on matrix Lie groups. The filter is used when the AUV is moving on the sea surface and it provides an estimate of the heading offset by comparing the speed measurements acquired by the Global Positioning System (GPS) and the Doppler Velocity Log (DVL) and by fusing the data coming from the IMU and the Fiber Optic Gyroscope (FOG). The initialization procedure has been validated with a dataset acquired by FeelHippo AUV in Cecina, Italy (September 2021).
{"title":"EKF on Lie Groups for Autonomous Underwater Vehicles orientation initialization in presence of magnetic disturbances","authors":"Alessandro Bucci, Leonardo Zacchini, A. Ridolfi","doi":"10.1109/AUV53081.2022.9965905","DOIUrl":"https://doi.org/10.1109/AUV53081.2022.9965905","url":null,"abstract":"Orientation estimation is a fundamental aspect of navigation and motion control of Autonomous Underwater Vehicles (AUVs). This concept is especially true when position sensors are unavailable and, consequently, navigation and control rely on dead reckoning strategies; in this case, orientation estimation is used in conjunction with speed measurements to update position estimation. When unknown magnetic disturbances are present, the magnetometers of the Inertial Measurement Unit (IMU) are unusable and do not provide an accurate initialization of the vehicle heading angle. This issue can be faced by applying a generalization of the Extended Kalman Filter in which the system state and measurements evolve on matrix Lie groups. The filter is used when the AUV is moving on the sea surface and it provides an estimate of the heading offset by comparing the speed measurements acquired by the Global Positioning System (GPS) and the Doppler Velocity Log (DVL) and by fusing the data coming from the IMU and the Fiber Optic Gyroscope (FOG). The initialization procedure has been validated with a dataset acquired by FeelHippo AUV in Cecina, Italy (September 2021).","PeriodicalId":148195,"journal":{"name":"2022 IEEE/OES Autonomous Underwater Vehicles Symposium (AUV)","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116343150","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-19DOI: 10.1109/AUV53081.2022.9965867
Pablo A. Gutiérrez-Flores, R. Bachmayer
The use of compact underwater vehicles for deep see exploration is still a big challenge in terms of available energy, reliability and robustness. Due to limited payload capacity these vehicles have to be equipped with narrow mission specific hardware. At the same time those vehicles still have to provide suitable navigation as well as actuation and communication solutions not unlike larger more capable vehicles. Using small compact vehicles in the deep sea requires different setups or at least a rapidly reconfigurable system that shares common building blocks in hardware and software in order to perform tasks. To this end this paper addresses the concept and development of a ROS2 based modular soft-and hardware architecture, which allows to decentralize and distribute different tasks by using microcontroller equipped modules in order to take advantage of a distributed data communication framework such as DDS (Data Distribution Service) and thus implement the microcontroller-oriented operating system (micro-ROS) in conjunction with ROS2 in the marine robotics domain. We report on initial tests and sea evaluations and consequently present an outlook toward the implementation of a new class of AUVs.
{"title":"Concept development of a modular system for marine applications using ROS2 and micro-ROS","authors":"Pablo A. Gutiérrez-Flores, R. Bachmayer","doi":"10.1109/AUV53081.2022.9965867","DOIUrl":"https://doi.org/10.1109/AUV53081.2022.9965867","url":null,"abstract":"The use of compact underwater vehicles for deep see exploration is still a big challenge in terms of available energy, reliability and robustness. Due to limited payload capacity these vehicles have to be equipped with narrow mission specific hardware. At the same time those vehicles still have to provide suitable navigation as well as actuation and communication solutions not unlike larger more capable vehicles. Using small compact vehicles in the deep sea requires different setups or at least a rapidly reconfigurable system that shares common building blocks in hardware and software in order to perform tasks. To this end this paper addresses the concept and development of a ROS2 based modular soft-and hardware architecture, which allows to decentralize and distribute different tasks by using microcontroller equipped modules in order to take advantage of a distributed data communication framework such as DDS (Data Distribution Service) and thus implement the microcontroller-oriented operating system (micro-ROS) in conjunction with ROS2 in the marine robotics domain. We report on initial tests and sea evaluations and consequently present an outlook toward the implementation of a new class of AUVs.","PeriodicalId":148195,"journal":{"name":"2022 IEEE/OES Autonomous Underwater Vehicles Symposium (AUV)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127102458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-06DOI: 10.1109/AUV53081.2022.9965808
Mabel M. Zhang, Woen-Sug Choi, Jessica Herman, D. Davis, Carson Vogt, Michael McCarrin, Yadunund Vijay, Dharini Dutia, William Lew, Steven C. Peters, B. Bingham
We present DAVE Aquatic Virtual Environment (DAVE)1, an open source simulation stack for underwater robots, sensors, and environments. Conventional robotics simulators are not designed to address unique challenges that come with the marine environment, including but not limited to environment conditions that vary spatially and temporally, impaired or challenging perception, and the unavailability of data in a generally unexplored environment. Given the variety of sensors and platforms, wheels are often reinvented for specific use cases that inevitably resist wider adoption.Building on existing simulators, we provide a framework to help speed up the development and evaluation of algorithms that would otherwise require expensive and time-consuming operations at sea. The framework includes basic building blocks (e.g., new vehicles, water-tracking Doppler Velocity Logger, physics-based multibeam sonar) as well as development tools (e.g., dynamic bathymetry spawning, ocean currents), which allows the user to focus on methodology rather than software infrastructure. We demonstrate usage through example scenarios, bathymetric data import, user interfaces for data inspection and motion planning for manipulation, and visualizations.1DAVE is available at https://github.com/Field-Robotics-Lab/dave
{"title":"DAVE Aquatic Virtual Environment: Toward a General Underwater Robotics Simulator","authors":"Mabel M. Zhang, Woen-Sug Choi, Jessica Herman, D. Davis, Carson Vogt, Michael McCarrin, Yadunund Vijay, Dharini Dutia, William Lew, Steven C. Peters, B. Bingham","doi":"10.1109/AUV53081.2022.9965808","DOIUrl":"https://doi.org/10.1109/AUV53081.2022.9965808","url":null,"abstract":"We present DAVE Aquatic Virtual Environment (DAVE)1, an open source simulation stack for underwater robots, sensors, and environments. Conventional robotics simulators are not designed to address unique challenges that come with the marine environment, including but not limited to environment conditions that vary spatially and temporally, impaired or challenging perception, and the unavailability of data in a generally unexplored environment. Given the variety of sensors and platforms, wheels are often reinvented for specific use cases that inevitably resist wider adoption.Building on existing simulators, we provide a framework to help speed up the development and evaluation of algorithms that would otherwise require expensive and time-consuming operations at sea. The framework includes basic building blocks (e.g., new vehicles, water-tracking Doppler Velocity Logger, physics-based multibeam sonar) as well as development tools (e.g., dynamic bathymetry spawning, ocean currents), which allows the user to focus on methodology rather than software infrastructure. We demonstrate usage through example scenarios, bathymetric data import, user interfaces for data inspection and motion planning for manipulation, and visualizations.1DAVE is available at https://github.com/Field-Robotics-Lab/dave","PeriodicalId":148195,"journal":{"name":"2022 IEEE/OES Autonomous Underwater Vehicles Symposium (AUV)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116794383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-15DOI: 10.1109/AUV53081.2022.9965917
Yiping Xie, Nils Bore, John Folkesson
Recent advances in differentiable rendering, which allow calculating the gradients of 2D pixel values with respect to 3D object models, can be applied to estimation of the model parameters by gradient-based optimization with only 2D supervision. It is easy to incorporate deep neural networks into such an optimization pipeline, allowing the leveraging of deep learning techniques. This also largely reduces the requirement for collecting and annotating 3D data, which is very difficult for applications, for example when constructing geometry from 2D sensors. In this work, we propose a differentiable renderer for sidescan sonar imagery. We further demonstrate its ability to solve the inverse problem of directly reconstructing a 3D seafloor mesh from only 2D sidescan sonar data.
{"title":"Towards Differentiable Rendering for Sidescan Sonar Imagery","authors":"Yiping Xie, Nils Bore, John Folkesson","doi":"10.1109/AUV53081.2022.9965917","DOIUrl":"https://doi.org/10.1109/AUV53081.2022.9965917","url":null,"abstract":"Recent advances in differentiable rendering, which allow calculating the gradients of 2D pixel values with respect to 3D object models, can be applied to estimation of the model parameters by gradient-based optimization with only 2D supervision. It is easy to incorporate deep neural networks into such an optimization pipeline, allowing the leveraging of deep learning techniques. This also largely reduces the requirement for collecting and annotating 3D data, which is very difficult for applications, for example when constructing geometry from 2D sensors. In this work, we propose a differentiable renderer for sidescan sonar imagery. We further demonstrate its ability to solve the inverse problem of directly reconstructing a 3D seafloor mesh from only 2D sidescan sonar data.","PeriodicalId":148195,"journal":{"name":"2022 IEEE/OES Autonomous Underwater Vehicles Symposium (AUV)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131019149","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：