WHOI's Autonomous Robotics and Perception Laboratory

WHOI Caribbean Reef Fish Detection Dataset

Levi Cai*, Austin Greene*, Nadège Aoki, Sierra Jarriel, Daniel Yang, T. Aran Mooney, Yogesh Girdhar Massachusetts Institute of Technology, Woods Hole Oceanographic Institution We introduce the WHOI Reef Solutions Initiative dataset for detection and classification of Caribbean Reef Fish. The dataset consists of 162

robotics

SeaSplat: Representing Underwater Scenes with 3D Gaussian Splatting and a Physically Grounded Image Formation Model

We introduce SeaSplat, a method to enable real-time rendering of underwater scenes leveraging recent advances in 3D radiance fields. Underwater scenes are challenging visual environments, as rendering through a medium

ecology

Workshop: Robots for Understanding Natural Ecosystems

We are excited to announce a new workshop titled "Robots for Understanding Natural Ecosystems" that we're hosting at the upcoming ICRA 2024, in Yokohama, Japan on May 17th! The workshop

publication

DeepSeeColor

Successful applications of complex vision-based behaviours underwater have lagged behind progress in terrestrial and aerial domains. This is largely due to the degraded image quality resulting from the physical phenomena

Projects

CUREE: A Curious Robot for Ecosystem Exploration

Coral reefs worldwide are threatened by anthropogenic disturbances and climate change. New tools are needed to scale up monitoring of coral reefs to understand reef ecosystems, rapidly assess biodiversity, and

publication

Semi-supervised Visual Tracking of Marine Animals Using Autonomous Underwater Vehicles

VMAT Dataset link Supplementary videos link In-situ visual observations of marine organisms is crucial to developing behavioural understandings and their relations to their surrounding ecosystem. Typically, these observations are collected

Co-Robotic Exploration

A day in the field, testing CUREE

WARPLab is in St. Johns, USVI from July 25 - Aug 7 2022. The focus of this op is to test new WARPAUV capabilities including audio-visual benthic surveys of coral

expedition

Field Trip to St. John, US Virgin Islands

We conducted robot field trials in USVI from Oct 17-Oct 30, 2021. The goal of the mission was to test our new robots, new robot behaviors, deploy acoustic sensors, and

Marine Animal Tracking Results

We presented a novel marine animal tracking dataset with evaluations of semi-supervised trackers as a poster, "Evaluation of Semi-supervised Methods for In-situ, Visual Tracking of Marine Animals", at the CV4Animals

Co-Robotic Exploration

Gaussian-Dirichlet Random Fields for Inference over High Dimensional Categorical Observations

Categorical observations, such as observations of phytoplankton taxa, are factored into the product of a community model and spatiotemporal distributions for each community. The community model, which is the distribution

Co-Robotic Exploration

Active Reward Learning for Co-Robotic Vision-Based Exploration in Bandwidth Limited Environments

Proposed approach to co-robotic exploration that models the interest of the operator over a low bandwidth communication channel and uses the learned reward model to plan the most rewarding (in

Co-Robotic Exploration

Information-Guided Robotic Maximum Seek-and-Sample in Partially Observable Continuous Environments

We present Plume Localization under Uncertainty using Maximum-ValuE information and Search (PLUMES), a planner for localizing and collecting samples at the global maximum of an a priori unknown and partially

Co-Robotic Exploration

ICRA2019: Streaming Scene Maps for Co-Robotic Exploration in Bandwidth Limited Environments

Abstract This work proposes a bandwidth tunable technique for real-time probabilistic scene modeling and mapping to enable co-robotic exploration in communication constrained environments such as the deep sea. The parameters

Co-Robotic Exploration

Barbados 2019 Field Trials

We just successfully completed our 2019 robot field trials at the Bellairs Research Institute 2019 Sea Trials in Holetown, Barbados. The main goals of the trials were to test our

Co-Robotic Exploration

IROS2018: Approximate Distributed Spatiotemporal Topic Models for Multi-Robot Terrain Characterization.

Our paper on enabling distributed learning in bandwidth limited environments was one of the finalists for the best paper award at IROS 2018 (6 finalists among 1,254 accepted). Abstract:

video

Film: Coding Curiosity

A short film by Laura Castanon et. al, describing the exploration robotics research happening in WARPLab.

Projects

Co-Robotic Exploration in Bandwidth Constrained Environments

This project addresses the control and communications among underwater robotic vehicles to explore and map in ocean environments, where the communications are inherently low bandwidth, may be degraded and even

papers

Learning Seasonal Phytoplankton Communities with Topic Models [2nd prize in the student poster competition at the OCEANS17]

We have developed a probabilistic generative model for phytoplankton communities. The proposed model takes counts of a set of phytoplankton taxa in a timeseries as its training data, and models

Summer Intern Project: SlickLizard Autonomous Surface Vehicle

USBL systems work best when there is not much horizontal distance between the transceiver and transponder. However, when deploying ROVs off a dock, there can be lots of distance between

Co-Robotic Exploration

IROS 2017: Feature discovery and visualization of robot mission data using convolutional autoencoders and Bayesian nonparametric topic modeling

Abstract: We introduce a novel unsupervised machine learning framework that incorporates the ability of convolutional autoencoders to discover features from images that directly encode spatial information, within a Bayesian nonparametric

papers

ICRA 2017: Phytoplankton Hotspot Prediction With an Unsupervised Spatial Community Model

Many interesting natural phenomena are sparsely distributed and discrete. Locating the hotspots of such sparsely distributed phenomena is often difficult because their density gradient is likely to be very noisy.

Visualizing AUV missions to Hannibal seamount

The following visualizations correspond to 16 dives by the SeaBed AUV at Hannibal seamount in Panama in 2015. These visualizations demonstrate that the learned model is automatically able to characterize

Field Trip: Using robots for characterizing coral reef soundscapes in US Virgin Islands

We recently went to US Virgin Islands to do some preliminary data collection of coral reef soundscapes using a robot. This trip was also our first attempt at deploying a

Projects

Robot Systems

CUREE: A Curious Underwater Robot for Ecosystem ExplorationCUREE: A Curious Underwater Robot for Ecosystem Exploration Robots with alternative locomotion design can assist us in observing previously inaccessible environments. MARE: Marine

Projects

Sensing the world, abstractly.

We want our robots to sense the world at varying levels of abstraction. From raw sensor value, to terrain type, to habitat type. ROST[1] is a technique for semantic