Lessons Learned from Designing and Evaluating a Robot-assisted Feeding System for Out-of-lab Use
This spotlight video provides an overview of our HRI '25 paper, which details our system design and out-of-lab evaluations, including a 5-day in-home deployment. See the associated publication for more details.
In-home Deployment of a Robot-assisted Feeding System for People with Motor Impairments
This video shows footage from a 5-day deployment of our robot-assisted feeding system in an end-user's home. He used it to feed himself meals while watching TV, while working, while in-bed, and more. See the associated publication for more details.
Multi-user, Out-of-lab Study of a Robot-assisted Feeding System for People with Motor Impairments
This video shows footage from an out-of-lab evaluation of our robot-assisted feeding system, where 6 users used the robot to feed themselves meals of their choice in an office, conference room, or public cafeteria. See the associated publication for more details.
Design Principles for Robot-Assisted Feeding in Social Contexts
This video presents our investigation into the challenges people with motor impairments face during social dining and how we can design a robot-assisted feeding system to address some of those challenges. See the associated publication for more details.
Unintended Failures of Robot-Assisted Feeding in Social Contexts
This humorous video highlights a variety of unintended consequences that can arise when a robot-assisted feeding system developed primarily for individual contexts is used in social contexts. This video aims to raise awareness about the importance of accounting for social context when designing assistive robots. See the associated publication for more details.
Evaluating an Assistive-Feeding Robot with Users with Mobility Limitations
In this work, we explore user preferences for different modes
of autonomy for robot-assisted feeding given perceived error
risks and also analyze the effect of input modalities on
technology acceptance.
Specifically, we tested: Speed(Fast vs. Slow), Interface
(Web-based vs. Voice-based), Environment (Social vs.
Individual), Level of Autonomy: (Full vs. Partial vs. Low)
Online Learning for Food Manipulation
This video shows how a robot can learn to skewer previously-unseen food items with different action distributions using online learning with a contextual bandit formulation.
Generalizing Skewering Strategies across Food Items
This video summarizes our work on SPANet framework and shows demonstrations of the robot’s skewering trials generalized across food items.
Transfer depends on Acquisition: Analyzing Manipulation Strategies for Robotic Feeding
Bite Acquisition with Tactile Sensing
The video shows calibration of FingerVision and Fingertip GelSight sensors for food manipulation application. The sensitivity of the sensors increase with low gripping force. In addition, it shows a control policy using which a robot can change the range and sensitivity of these tactile sensors by controlling gripping forces.