Search for: All records

Robots should personalize how they perform tasks to match the needs of individual human users. Today’s robots achieve this personalization by asking for the human’s feedback in the task space. For example, an autonomous car might show the human two different ways to decelerate at stoplights, and ask the human which of these motions they prefer. This current approach to personalization isindirect: Based on the behaviors the human selects (e.g., decelerating slowly), the robot tries to infer their underlying preference (e.g., defensive driving). By contrast, our article develops a learning and interface-based approach that enables humans todirectlyindicate their desired style. We do this by learning an abstract, low-dimensional, and continuous canonical space from human demonstration data. Each point in the canonical space corresponds to a different style (e.g., defensive or aggressive driving), and users can directly personalize the robot’s behavior by simply clicking on a point. Given the human’s selection, the robot then decodes this canonical style across each task in the dataset—e.g., if the human selects a defensive style, the autonomous car personalizes its behavior to drive defensively when decelerating, passing other cars, or merging onto highways. We refer to our resulting approach as PECAN:Personalizing Robot Behaviors through a LearnedCanonical Space. Our simulations and user studies suggest that humans prefer using PECAN to directly personalize robot behavior (particularly when those users become familiar with PECAN), and that users find the learned canonical space to be intuitive and consistent. See videos here:https://youtu.be/wRJpyr23PKI. 

This folder contains  model extractions for the manuscript "The Limits of Oceanic Forcing on the Exchange Flow of the Salish Sea" by Robert Sanchez, Sarah Giddings, and Emily Lemagie. At the moment, data is still being uploaded. The model extractions contain hourly fields of ssh, wind stress, u,v,temp,and salt. Feel free to reach out to the corresponding author (Robert Sanchez) for code or clarifications. Code to produce TEF from these sections will be available on GitHub soon. The data comes from the LiveOcean model (here), with other extractions available here. Model details can be found here. 

Abstract Freshwater from the Greenland Ice Sheet is routed to the ocean through narrow fjords along the coastline where it impacts ecosystems both within the fjord and on the continental shelf, regional circulation, and potentially the global overturning circulation. However, the timing of freshwater export is sensitive to the residence time of waters within glacial fjords. Here, we present evidence of seasonal freshwater storage in a tidewater glacial fjord using hydrographic and velocity data collected over 10 days during the summers of 2012 and 2013 in Saqqarleq (SQ), a midsize fjord in west Greenland. The data revealed a rapid freshening trend of −0.05 ± 0.01 and −0.04 ± 0.01 g kg −1 day −1 in 2012 and 2013, respectively, within the intermediate layer of the fjord (15–100 m) less than 2.5 km from the glacier terminus. The freshening trend is driven, in part, by the downward mixing of outflowing glacially modified water near the surface and increasingly stratifies the fjord from the surface downward over the summer melt season. We construct a box model that recreates the first-order dynamics of the fjord and describes freshwater storage as a balance between friction and density-driven exchange outside the fjord. The model can be used to diagnose the time scale for this balance to be reached, and for SQ we find a month lag between subglacial meltwater discharge and net freshwater export. These results indicate a fjord-induced delay in freshwater export to the ocean that should be represented in large-scale models seeking to understand the impact of Greenland freshwater on the regional climate system.