Expert Profile

Yu Sun

Professor

COMPUTER SCIENCE AND ENGINEERING | COLLEGE OF ENGINEERING

Research Terms & Keywords

Research Terms

Computer Science Artificial Intelligence Computer Vision Robotics

Research Projects

Ends

Title

07-2029

Developing AI-Driven Pain Intensity and Pain Sensitizati Developing AI-Driven Pain Intensity and Pain Sensitizati

06-2026

POSE: Phase II: Collaborative Open-source Manipulation P POSE: Phase II: Collaborative Open-source Manipulation P

06-2023

An AI-based multimodal approach to predict Pain in postn An AI-based multimodal approach to predict Pain in postn

06-2022

RI: Small: Generalizing Learned Manipulation Skills to Unseen Situations By Balancing Uncertainties RI: Small: Generalizing Learned Manipulation Skills to Unseen Situations By Balancing Uncertainties

01-2022

A Multimodal Approach for Monitoring Prolonged Acute Pai A Multimodal Approach for Monitoring Prolonged Acute Pai

06-2021

CHS: Small: Collaborative Research: Wearable Fingertip H CHS: Small: Collaborative Research: Wearable Fingertip H

12-2020

REU Site on Ubiquitous Sensing REU Site on Ubiquitous Sensing

08-2020

Microsoft Research Dissertation Grant -- Searching High Microsoft Research Dissertation Grant -- Searching High

04-2019

An Automated Pressure Ulcer Monitoring System to Improve An Automated Pressure Ulcer Monitoring System to Improve

09-2018

Developing Deep Neural Networks to Detect Objects and Th Developing Deep Neural Networks to Detect Objects and Th

02-2018

EAGER: Characterizing Physical Interaction in Instrument EAGER: Characterizing Physical Interaction in Instrument

07-2017

RI: Small: Functional Object-Oriented Network for Manipu RI: Small: Functional Object-Oriented Network for Manipu

08-2011

CPS: Small: Virtually Transparent Epidermal Imagery CPS: Small: Virtually Transparent Epidermal Imagery

Degrees

Ph.D., Computer Science

University of Utah ▪ 2007

Publications

Professional Presentations

Task-Oriented Grasp Planning Based on Disturbance Distribution; ISRR; ISRR; 2013

Grasp Planning Based on Grasp Strategy Extraction from Demonstration; IEEE/RSJ International Conference on Intelligent Robots and Systems; IEEE; 2014

Functional Analysis of Grasping Motion; IEEE/RSJ International Conference on Intelligent Robots and Systems; IEEE; 2013

Grasp Mapping Using Locality Preserving Projections and KNN Regression; IEEE Intl. Conference on Robotics and Automation; IEEE; 2013

Human-Object-Object-Interaction Affordance; Workshop in Robot Vision (WoRV)/Winter Vision Meeting ; IEEE; 2013

Five Most Important Trends in Robotics Research; Link: Hardware Partners @ Silicon Valley; Makerlink; 2015

Determining the Benefit of Human Input in Human-in-the-Loop Robotic Systems; IEEE ROMAN ; IEEE; 2013

Virtually Transparent Epidermal Imagery; NSF CPS PI Meeting; NSF; 2013

From Knowledge to Action-- Toward Robotic Cooking; ICRA Workshop on Sensor-Based Object Manipulation for Collaborative Assembly; International Conference on Robotics and Automation; 2017

Hand and Mind; Singularity University; 2016

Ideomotor Learning for Robotic Manipulation; Bay Area Robotics Symposium; Stanford University; 2016

Bring AI into Physical World through Robotic Hands; Google; 2016

Renaissance of Robotic Grasping; CCF Global Artificial Intelligence & Robotics Summit; Chinese Computing Foundation; 2016

Tasks in Robotic Grasping and Manipulation Competitions; IROS Workshop on Development of Benchmarking Protocols for Robot Manipulation; IROS; 2017

Other Professional Activities

Co-Chair, Member Services Committee, Member Activities Board Board Member, Electronic Products and Services Board Associate Editor, IEEE Robotics & Automation Magazine Associate Editor, , IEEE International Conference on Robotics and Automation (ICRA) 2014 Editorial Advisory Board of Assembly Automation journal

Stories

Using AI to Detect Silent Pain in Newborns

#PainInTheNICU

Center

Center for Innovation, Technology, and Aging

Director	Yu Sun
Phone
Website	https://www.usf.edu/engineering/cita/index.aspx
Mission	CITA is dedicated to transforming the landscape of care for individuals with AD/ADRD, and subsequently for other vulnerable older adults with disabilities such as Parkinson's disease and those recovering from stroke. To achieve this, we take a two-pronged approach that considers both patients and their caregivers. CITA aims to innovate and create technologies and solutions for extending safe stay at home and for receiving timely and high-quality care that is cost-effective and customized to meet the unique needs of ADRD individuals. In addition, we will design innovative technologies to enhance the capabilities of caregivers, significantly reducing their burden while supporting their wellbeing and work-life balance. Looming demographic changes and the growing incidence of Alzheimer’s Disease and Related Dementia (AD/ADRD) present a formidable challenge to the U.S. and global healthcare systems. By 2030, the U.S. Census Bureau projects that 21% of the U.S. population will be over 65, outnumbering children for the first time in history. Further, according to the United Nations, by 2050, half of the world's population will reside in countries where 20% of the population is over 65. This demographic shift heralds a significant increase in the ADRD prevalence, with current figures at approximately 55 million people worldwide, expected to rise to 153 million by 2050. In the United States, over 6.2 million aged 65 and older are currently living with ADRD, a number projected to surge by 47% to 13.8 million by 2060. CITA has a team of 35 affiliated professors from eight colleges across the three USF campuses: College of AI, cybersecurity and computing; College of Engineering; College of Behavioral and Community Sciences; College of Arts and Sciences; Morsani College of Medicine; College of Education; College of Nursing; College of Public Health. The Center is supported by a team of research coordinators and research assistants. We continue to look for more affiliated professors and collaborators.

Technologies

Systems And Methods For Learning And Generating Movement Policies For A Dynamical System

Competitive Advantages: Adaptable learning, Flexibility in realistic situations, Makes real time decisions. System and methods for generating a trajectory of a dynamical system are described herein. An example method includes modelling a policy from demonstration data. The method also includes generating a first set of action particles by sampling from the policy, where each of the action particles in the first set includes a respective system action, predicting a respective outcome of the dynamical system in response to each of the action particles in the first set, and weighting each of the action particles in the first set according to a respective probability of achieving a desired outcome. The method further includes generating a second set of action particles by sampling from the weighted action particles, where each of the action particles in the second set includes a respective system action, and selecting a next system action in the trajectory of the dynamical system from the action particles in the second set.USF inventors have created a trajectory generation approach that learns a broad and stochastic policy from demonstrations. This approach first learns from the human demonstrations of the policies and is able to generate a trajectory with arbitrary length, and treats the change of the constraints naturally. Based on the magnitude of action and the desired future state, the dynamic system decides to keep, reject or execute samples. The approach weighs the action particles drawn from the policy by their respective likelihood of achieving a desired future outcome, and obtains the optimal action using the weighed actions.