June 2022 OES Beacon

The Student Poster Competition at OCEANS 2022 Chennai

Dr. A Malarkodi, SPC Chair OCEANS 2022 Chennai, Dr. Shyam Madhusudhana, OES Student Poster Competition Chair

SPC participants from left First row – XiaoGang Li, Zhao Fan, Wang Yaomei, Sekimori Yuki, Mark Ali, Second row – Fabricio Bozzi, Cesar Rojas, Rashi Srivastava, Lady Nicole Macas Mendez, Subodh Bhosale, Third Row – Zhiding Yang, Paulo Padrao, Yang Weng, Tata Nandhini, Sai Ganesh (no photos for Ashutosh Rastogi and HaoDong Qi)

The Student Poster Competition (SPC) is a flagship event of the MTS/OES OCEANS conferences in which undergraduate and graduate students from colleges and universities around the world participate. The SPC at OCENS 2022 Chennai received 52 submissions from around the globe, out of which 19 were selected for the participation in the final program following rigorous reviews by professionals in the field (finally, 17 students participated). This edition of the SPC was financially supported by      a generous grant      from      the Office of Naval Research-Global. The      monetary awards      for this SPC were      supported      by Schmidt Ocean Institute and National Oceanic and Atmospheric Administration.      We thank the sponsors for their continued support.

SPC Jury members, LOC members and volunteers with Chair

All selected participants were asked to present their posters in three sessions online. First two sessions were conducted on the 22nd of February, 2022. In this session, 11 posters were presented out of 13. On the 23rd of February, 2022, 5 posters were presented out of 6 posters including two in-person      participants. We had a      panel of eminent juries consisting of Professor V. Sundar, Emeritus professor as chair along with      Prof.      Asokan Thondiath, Dr. Vedachalam and Dr. Muthuvel     . Winning participants      were announced during the Gala Dinner on the 23rd of February, 2022. The prize money associated with the awards were      arranged to be      sent to the      winners by Schmidt Ocean Institute through MTS. This year, a new award category was introduced —           the top      scored poster from among the host country participants was chosen      and a special prize,      sponsored by LOC, OCEANS 2022 Chennai, was awarded.

SPC participants with their affiliation, poster title and abstracts are given below.

First prize (Norman Miller Award) (Certificate and $ 3000)

Yang Weng, The University of Tokyo, Tokyo, Japan

Sim-to-Real Transfer for Underwater Wireless Optical Communication Alignment Policy between AUVs

 Abstract—The underwater wireless optical communication (UWOC) technology provides a potential high data rate solution for information sharing between multiple autonomous underwater vehicles (AUVs). In order to deploy the UWOC system on mobile platforms, It is proposed to solve the optical beam alignment problem by maintaining the relative position and orientation of two AUVs. A reinforcement learning based alignment policy is transferred to the real world since it outperforms other baseline approaches and shows good performance in the simulation environment. We randomize the simulator and introduce the disturbances, aiming to cover the real distribution of the underwater environment. Soft actor-critic (SAC) algorithm, reward shaping based curriculum learning, and specifications of the vehicles are utilized to achieve the successful transfer. In the Hiratsuka sea experiments, the alignment policy was deployed on the AUV TriTON and successfully aligned with autonomous surface vehicle BUTTORI. It demonstrates a solution for combining the UWOC technology and AUVs team in the ocean investigation.

Second Prize (Certificate and $2000)

Zhiding Yang, Memorial University of Newfoundland, Canada

A Temporal Convolution Network for Wave Height Estimation from X-band Radar Data

 Abstract—A state-of-the-art machine learning based significant wave height (Hs) estimation model, which is based on a Temporal Convolution Network (TCN), is proposed for X-band marine radar in this paper. The input space of the network is composed of three features (i.e., signal-to-noise ratio (SNR)-based, ensemble empirical mode decomposition (EEMD)-based, and gray level concurrence matrix based features) extracted from radar images. Two typical Hs estimation methods (i.e., SNR-based linear regression and EEMD-based linear regression methods) are utilized for comparison with the proposed method using the radar and buoy data collected at the East coast of Canada. It is found that the proposed method can generate the most accurate Hs results with a root-mean-square error of 0.24 m and a correlation coefficient of 0.94.

Third Prize (Certificate and $1000)

Fabricio A. Bozzi, University of Algarve, Portugal

Vector Hydrophone Passive Time Reversal for Underwater Acoustic Communications

Abstract—The use of vector hydrophones as a receiver for underwater communications has been the subject of research since such a device is a compact option to pressure-only arrays. A vector hydrophone, usually called acoustic vector sensor, is a device that measures pressure and particle velocity components. This paper investigates a method to combine those channels based on passive time-reversal (PTR). Simulation and experimental data are used to quantify communication performance, comparing vector hydrophones to pressure-only arrays. The analyzed acoustic scenario consists of a shallow-water area (about 100 m), where a vector hydrophone array receives communication signals from a bottom moored source. Simulations help in the understanding of diversity by analyzing spectral characteristics of vector hydrophone channels and the PTR q-function. While in simulation, the benefits of PTR using particle velocity channels are perceptible seen by exploring diversity, communication performance with experimental data is degraded due to time varying. Finally, the achieved performance using a single or a small array of vector hydrophones enforces its benefits for communication enhancement.

Special Prize (Certificate and INR 25,000)

Tata Nandini, Sri Sai Ram Engineering College, Chennai, India

Real-Time classification of Plankton species using Convolutional Neural Networks

 Abstract— Marine geology involves investigation of marine species for ocean observations. Marine geologists study the history, the processes occurring at the ocean floor to derive valuable insights about marine life. This study plays a vital role in maintaining balance in our ecosystem. The marine ecosystem houses species ranging from energy producers such as aquatic plants, phytoplanktons and consumers like fishes to humans. The study of one such marine species called “plankton” is known as planktology. Planktons are microscopic species that play a rather unnoticed, but a pivotal role in the marine food web. This leads to a need for a proper identification system which classifies different plankton species and also has a record of them which further helps in recognizing new plankton species. This system can also be used as a tool for classification of a particular plankton species with the help of Convolutional Neural Networks, which is a tedious process otherwise. This model is further deployed as a web app which can also allow authenticated users to contribute to the data collection of the model. Thus, this system will have a major impact on knowing different plankton species and also makes the subject of planktology more feasible.

Ashutosh Rastogi, Indian Institute of Technology New Delhi, India

Cavitation visualization and prediction of propeller characteristics of INSEAN E779A propeller using sliding mesh model

Abstract— A Marine propeller is an ensemble of airfoil sections assembled in the form of a propeller blade. The pressure in the liquid adjoining to the body drops as the square of the local flow velocity as the propeller rotates. While the propeller rotates in water, the difference in stresses existing between the face and back of the propeller creates a thrust force in the forward direction, which lets it overcome the drag experienced by the vessel. Different methods are available to determine propeller performance, including experimental, theoretical and numerical. The open water propeller tests are conducted either in towing tanks/ circulating water channels to determine the non dimensional propeller parameters such as Thrust co-efficient and Torque co-efficient and though accurate in the model scale are cumbersome and costly. The Computational Fluid Dynamics (CFD) techniques provide a reliable and robust solution to this problem. The current work presents the numerical prediction method to determine hydrodynamic performance characteristics of an INSEAN E779A propeller. The study is implemented using the commercially available computational fluid dynamics (CFD) solver, Ansys Fluent. The study utilized unstructured tetrahedral meshing along with wedge/prism-shaped elements in the boundary layer regions, with a Realizable k-ɛ turbulence model. The sliding mesh model is used to replicate the physics of the rotating propeller. The study has been carried out using Unsteady RANS. The solver employed is SIMPLE. Studies have been carried out using both first order upwind and second order upwind for pressure, momentum and transient formulation. However, it is seen that both the formulations give the same accuracy. A detailed study of time step estimation has been carried out to arrive at the most optimum time step without sacrificing the accuracy of the results. To capture the cavitation effect, the phases of water and air were formulated using the Mixture option in Fluent with Mass Transfer mechanism at cavitation and cavitation model as Schenerr-Sauer model. Results show reliable thrust coefficient, torque coefficient, and efficiency data for the case of low advance ratios. The cavitation is visualized by iso-surfaces of pressure below the vapour pressure and was validated by available photographs from experiments conducted in a cavitation tunnel with “reasonable matching.”

Rashi Srivastava, Central University of Karnataka, Karnataka, India

Analysis of Sea Waves and Ship Wake Detection

 Abstract—Surface has been of great importance to the researchers seeking to understand the sea waves with the aim of benefitting the safe undertaking of various maritime activities. Synthetic Aperture Radar (SAR) images carry a lot of such information and characteristic patterns which can be exploited to understand sea surface currents and ship behaviors. Through this work, we have studied sea wave patterns and the formation of ship wakes. To retrieve the direction of sea currents from these radar images, algorithms based on 2D Fast Fourier Transform (Fast Fourier Transform) has been employed. SAR images can be taken into consideration for ship route estimation as well by studying the behavior of ship wakes and ship movement. Ship wakes have a pivotal role in the analysis of SAR images of a sea due to the volume of information which can be extracted through them. In this work, we have explored Hough Transform as a line detection method in which linear features are enhanced. Several Hough transform techniques are implemented on these images to obtain and demarcate visible linear features formed by the ship wakes. Each image point is mapped to detect parameterized shapes; hence, it reduces the possibility of detecting spatially spread patterns in the image. For the detection procedure ships containing tiles from the original images are tiled out. The proposed work has been experimented on Setinel-1B SAR images with multiple ships present in them. The results obtained show a fairly accurate detection performance. This study has been conducted on reliable radar data and several features of the sea waves and surface have been explored.

Yuki Sekimori, The University of Tokyo, Tokyo, Japan

AUV ARIEL: Computer-Vision-Driven Intervention Processed on a Small Single-Board Computer

 Abstract—ARIEL, a newly developed small-sized, low-cost, and modulus hovering type autonomous underwater vehicle, is designed to perform computer-vision-driven intervention tasks on a small single-board computer. The working prototype has demonstrated an underwater intervention mission and proved a functional system at the Underwater Robot Convention in JAMSTEC 2021. In the experiment trials, ARIEL performed the mission in the testing water tank. ARIEL is a reference design of an AUV for lightweight shoreline intervention tasks.

Mark Ali, Jacobs University Bremen, Bremen, Germany.

Fault Detection in AUV navigation: a computationally inexpensive approach

 Abstract—This paper proposes a computationally inexpensive approach for fault detection during AUV navigation. Thruster performances may degrade over time, or a complete failure may occur. Considering that currently most of AUV missions are surveys involving long linear transits, this paper focuses on analysis of line segments and detection of navigation errors with varying thruster performances degradation, using linear approximation and curvature analysis. Those mathematical approaches have been chosen due to their computational efficiency, an important aspect for low-cost vehicles and for a service watch module, which needs to run in the background without competing for resources with mission-critical modules. Extensive tests have been run with the Sparus AUV at the University of Girona facilities and at sea, confirming the validity of the proposed approach.

Yaomei Wang, Memorial University of Newfoundland, Canada

An experimental study of the cooperation between sonar and a fluorometer for detecting underwater oil from an underwater vehicle

 Abstract— In oil spill events, using a combination of multiple types of sensors is favorable to improve the performance of oil detection and the reliability of data collected. An experiment was conducted to investigate the feasibility of using a fluorometer and a sonar to detect and

cross-validate the presence of spilled oil before being installed and used on a Slocum glider. Considering the limited payload capability and energy availability of a Slocum glider, a lightweight Cyclops submersible fluorometer and a Ping 360 sonar were used to cross-validate fluorescence measurements and sonar images. The results from the experiment conducted in a tank with oil release confirmed the feasibility of using both fluorometers and sonar to verify the existence of spilled oil.

Lady Nicole Macas Mendez, Faculty of Maritime Engineering and Marine Sciences, Ecuador

Investigation of Suitable Areas for Integrated Multi-Trophic Culture of Kappaphycus alvarezii and Crassostrea gigas in Santa Elena-Ecuador

 Abstract—According to the Food and Agricultural Organization the mariculture has generated interest in recent years due to its importance in obtaining foods of high nutritional value for the growing world population. In Ecuador, the aim is to promote the diversification of the aquaculture sector, so this project goals to identify potential areas for the multi-trophic cultivation of Kappaphycus alvarezii and Crassostrea gigas, in Santa Elena – Ecuador. To enhance the production of these species spatial planning is necessary, based on the biological requirements of the organisms to be cultivated, the current legal regulations and considering possible conflicts of use. Nine environmental and socio- economic criteria were identified and obtained, making use of Geographic Information Systems (GIS) and remote sensing. Subsequently, the reclassification of these criteria was carried out based on the optimal cultivation ranges of both species, to finally continue with the Boolean superposition of layers, resulting in a map showing the areas suitable for cultivation. These results indicate that multi-criteria analysis offers an overview that is important for decision-making within integrated coastal management and spatial planning.

Cesar A. Rojas, Florida International University, Miami, USA

Combining Remote and In-situ Sensing for Persistent Monitoring of Water Quality

 Abstract—Many studies suggest that water quality parameters can be estimated by applying statistical and machine learning methods using remote sensing or in-situ data. However, identifying best practices for implementing solutions appears to be done on a case-by-case basis. In our case, we have in-situ data that covers a large period, but only small areas of Biscayne Bay, Florida. In this paper, we combine available in-situ data with remote sensing data captured by Landsat 8 OLI-TIRS Collection 2 Level 2(L8), Sentinel-2 L2A(S2), and Sentinel-3 OLCI L1B(S3). The combined data set is for use in a water quality parameter estimation application. Our contributions are two-fold. First, we present a pipeline for data collection, processing, and co-location that results in a usable data set of combined remote sensing and in-situ data. Second, we propose a classification model using the combined data set to identify areas of interest for future data collection missions based on chlorophyll-a in-situ measurements. To further prove our methodology, we conduct a data collection mission using one of the predicted paths from our model.

Paulo Padrao, Florida International University, Miami, USA

Towards Learning Ocean Models for Long-term Navigation in Dynamic Environments

 Abstract—The use of underwater robot systems, including Autonomous Underwater Vehicles (AUVs), has been studied as an effective way of monitoring and exploring dynamic aquatic environments. Furthermore, advances in artificial intelligence techniques and computer processing led to a significant effort towards fully autonomous navigation and energy-efficient approaches. In this work, we formulate a reinforcement learning framework for long-term navigation of underwater vehicles in  dynamic environments using the techniques of tile coding and eligibility traces. Simulation results used actual oceanic data from the Regional Ocean Modeling System (ROMS) data set collected in Southern California Bight (SCB) region, California, USA.

Subodh Bhosale, Indian Institute of Technology Bombay, India.

Design of a Health Monitoring System for SCUBA Divers

 Abstract—This work proposes a novel diver health monitoring system (HMS) for continuously measuring cuffless blood pressure (BP) via exploiting the pulse arrive time (PAT) between electrocardiogram (ECG) and photoplethysmogram (PPG) signals. Pulse Arrival Time (PAT) is the time difference between ECG, R-peak and PPG peak from synchronized time signals. The proposed solution effectively correlates the PAT with BP and proves the same with data obtained from both normal and elevated BP patients. This system enables continuous, cuffless and minimally obtrusive measurement of BP on a continuous basis and can detect sudden increase in transient BP levels to raise an alert. This system can also serve as a platform for acquiring synchronized ECG and PPG signals for detecting various other cardiac ailments such as Arrhythmias, Tachycardia, etc.

Fan Zhao, The University of Tokyo, Tokyo, Japan

New method of mussel survey by using high resolution acoustic video camera-ARIS and deep learning

 Abstract—Due to water transparency, water depth, and higher labor demand, conventional methods for the underwater survey (e.g., optical sensing and quadrat survey) have their limitations. Thus, to overcome these barriers, this paper proposes a method of acoustic sensing which uses the high-resolution acoustic video camera-ARIS to visualize the lake bottom and investigate the distribution of mussels. The New underwater sensing method produces near-video quality acoustic images for constructing the map by Image Mosaic Operation, which can be helpful for assessing the status of mussels. Convolutional Neural Network(CNN) shows its help in the detection and classification of mussels in this study. Meanwhile, the accuracy and efficiency of the well-trained deep learning model manage to improve this research. Through the field survey, the proposed method successfully obtained the distribution maps of mussels in Lake Izunuma.

Sai Ganesh CS, Sri Sai Ram Engineering College Chennai, India

Machine Learning Based Classification and Modelling of Underwater Acoustic Communication

Abstract – The acoustic medium in the ocean has high complications due to its non-homogenous property. The speed of sound in the medium plays a significant role in acoustic computations and is more related to the density and compressibility of the propagation medium. Several acoustic propagation modeling methods that are described by wave equations are proposed for different underwater applications. The mathematical propagation models that are used widely are the empirical method (Thorp’s model), ray theory (Bellhop model), normal mode method (Kraken), wavenumber integration (Scooter), and parabolic equations (RAMGeo). The propagation models compute several parameters that include transmission loss, impulse response, arrival time, etc. with the input of the sound velocity profile and the transmission environment. The error rate of the propagation models varies with respect to the frequency, range of transmission and other parameters as well. In this paper, a classification dataset for shallow water propagation is generated with the threshold limits of range and frequency of each propagation model. Since, the limits of the propagation model are non-linear, machine learning based algorithms are proposed and validated with the data generated. Finally, a GUI is created that classifies the required propagation model and simulates the model with the inputs of range, frequency and sound velocity profiles.

XiaoGang Li, Ocean University of China, China

A new fuzzy SMC control approach to path tracking of autonomous underwater vehicles with mismatched disturbances

Abstract—This work proposed a new Sliding Mode Control (SMC) scheme for path tracking of autonomous underwater vehicles (AUV) with uncertainties and mismatched disturbances. Considering the complexity of parameter tuning, a fuzzy system based gain-scheduling scheme is designed to deal with the system tuning. In the control system, mismatched disturbances and uncertainties are estimated and assessed by disturbance observers (DOB) and Radial Basis Function Neural Networks (RBFNN), respectively. Additionally, the condition of Lyapunov stability is analyzed here to ensure the closed-loop stability of the control system. This control scheme can be applied to path tracking of AUV in different driving environments within fast changing trajectories. Finally, many comparative experiments have been conducted to show the favorable performance of the proposed controller.

HaoDong Qi, Harbin Engineering University, China

Autonomous Underwater Rescue Technology

Abstract—This paper proposes an underwater autonomous rescue technology for drowning people, which is composed of an underwater rescue vehicle and rescue AI algorithm. The vehicle consists of a shell, power unit, and added buoyancy device, which can help the drowning person from the underwater to the surface, and the power device can drive the vehicle to move. The rescue AI algorithm consists of human body recognition AI technology based on deep learning and path planning methods based on improved PSO. Also, this paper carries out numerical analysis on-resistance of the vehicle, simulates the path planning based on PSO, and tests several functions in the lake.