MIT Logo

Event Category: Ocean Engineering Seminar

Swarm Locomotion in Time-Dependent Systems using Multi-Agent Reinforcement Learning

Room: 5-314 Speaker Name: Prof. Oren Gal Affiliation : Hatter Department of Marine Technologies, Leon H. Charney School of Marine Sciences, University of Haifa Abstract: In this talk, we harness advanced data science techniques, Multi-Agent Reinforcement Learning (MARL), and Computational Fluid Dynamics (CFD) to optimize micro-robotic swarm locomotion under time-dependent flow conditions. Traditional approaches often…

Learning Turbulent Flows with Generative Models: Super-resolution, Forecasting, and Sparse Flow Reconstruction

Room: 5-314 Speaker Name: Oommen, Vivek Affiliation: Brown University Abstract: Neural operators are promising surrogates for dynamical systems but when trained with standard L2 losses they tend to oversmooth fine-scale turbulent structures. Here, we show that combining operator learning with generative modeling overcomes this limitation. We consider three practical turbulent-flow challenges where conventional neural operators…

Vortex Detection Mechanisms of Harbor Seal Whiskers for Biomimetic Sensing

Room: 5-314 Speaker Name: Adrienne Lai Affiliation: PI: Michael Triantafyllou Abstract: Harbor seal whiskers have the characteristic ability to detect changes in flow velocity at the micrometer per second scale, allowing them to track prey with their whiskers as the primary sensor. This unique property has inspired researchers to design low-power, non-invasive, passive sensors to…

Applying Solar Cell Principles to Determine Ocean Wave Intensity in a Sea Ice Sheet in the Marginal Ice Zone

Room: 5-314 Speaker Name: Max Pierce Affiliation: PI: Dick Yue Abstract: Inspired by the well-known Yablonovitch limit which provides an upper bound on the light intensity enhancement in a solar cell, we derive a simple expression relating the ocean wave intensity in a floating ice sheet within the marginal ice zone (MIZ) to the wave…

Generative AI for weather data assimilation by differentiating through flows

Room: 5-314 Speaker Name: Ruizhe Huang Affiliation: PI: Sherrie Wang Abstract: To anchor weather products in reality, data assimilation integrates physical simulations of the atmosphere with observational data. Traditional approaches achieve this under assumptions of Gaussian errors and linearized dynamics, which limit accuracy. Deep generative models offer a flexible alternative, yet existing guidance-based approaches are…

R.B. Wallace Lecture: Increasing sustainable use of the oceans: perspective and challenges for ocean engineers

Room: 5-314 Speaker Name: David Kelly Affiliation: CEO of Innovasea Abstract: With an increasing global population, the utilization of the oceans to economically support societies is growing. While the traditional uses of the oceans for transportation, defense, energy, and food continue, the engineering challenges are evolving and the range of disciplines involved is increasing. This lecture…