The Michigan Robotics Newsletter, Fall 2022
Note: these posts were migrated from Twitter’s Revue newsletter after it shut down in January, 2023. Unfortunately links will not redirect at this time, but we hope to find some time to go through these and relink in the future.
Welcome to the Michigan Robotics Newsletter, a summary of what’s happening in the University of Michigan Robotics community.
The 2022 Robotics Outreach Ambassadors
Students are the largest part of our robotics community, and regularly take this outreach task on in addition to their coursework, research, student instruction workloads. While there are many, many students who take part in volunteering, there are several that have gone above and beyond in the past year.
Research
‘Fake’ data helps robots learn the ropes faster
In a step toward robots that can learn on the fly like humans do, Peter Mitrano and Dmitry Berenson provide a new approach expands training data sets for robots that work with soft objects like ropes and fabrics, or in cluttered environments.
How we can better link mind and machine
Man I (Maggie) Wu and Leia Stirling tested how exoskeleton users responded to the task of matching haptic feedback to the timing of each footstep, finding added mental workload, causing less effective use of the exoskeleton.
Invariant Kalman Filter for wheeled robot
From Maani Ghaffari‘s CURLY Lab, the InEKF takes in IMU measurements and body velocity estimation and estimates the robot’s pose, velocity, and IMU biases
Cross-cultural investigation of drivers’ trust & preferences in AVs
Na Du, Lionel Robert, and X. Jessie Yang reveal that people in high-context cultures have higher trust in and preference for automated vehicles.
Variational inference MPC using normalizing flows and out-of-distribution projection
What happens when a learning-based Model Predictive Control algorithm hallucinates an in-distribution environment for generating low-cost trajectories? Thomas Power and Dmitry Berenson answer.
Robot attractiveness and team identification on performance and viability in human–robot teams
Sangseok You and Lionel Robert show that team identification in human–robot teams leads to better performance and team viability.
Disengagement cause-and-effect relationships extraction using NLP
Yangtao Zhang, X. Jessie Yang, and Feng Zhou help us understand the causes of Autonomous Vehicle Disengagement to provide guidance for AV testing and deployment.
A literature review of trust repair in HRI
Connor Esterwood and Lionel Robert present a cohesive view of when apologies, denials, explanations, and promises have been seen to repair trust.
Effects of autonomy transparency on trust, dependence, and human-autonomy team performance
Ruikun Luo, Na Du and X. Jessie Yang find that with enhanced autonomy transparency, people’s trust increases significantly over time.
Effect of a powered ankle exoskeleton on agility with inertial measurement units
Katelyn King, Sarah Gonzalez, and Leia Stirling quantified agility through performance on turns while wearing unpowered and powered ankle exos.
openaccess.cms-conferences.org
Functional chemical motor coatings for powering of self-propelled particles
Abdon Pena-Francesch and Chia-Heng Lin led a group that demonstrated modular protein micromotors to power inactive objects of many sizes and designs.
Congrats
2022 Qualcomm Innovation Fellowship
Anthony Opipari and Alphonsus Adu-Bredu won for research in energy-efficient algorithms for power-constrained mobile robots.
Other congratulations
Michigan Mars Rover won the University Rover Challenge
Nima Fazeli, Maani Ghaffari, and Jean-Baptiste Jeannin received Amazon Research Awards
Ella Atkins, moving to a new role as Department Chair of Aerospace & Ocean Engineering at Virginia Tech
John Laird, on retiring after 36 years
Kira Barton‘s Research Group, in collaboration with Dawn Tilbury, won Manufacturing Leadership Award
Edwin Olson’s May Mobility closed $111M in Series C funding
Look
PhD Defense: UAS Contingency Management Autonomy with Experimentally Validated Models
Prashin Sharma presents research on multicopter flight performance and improving safety of small Uncrewed Aerial Systems.
PhD Defense: Non-parametric Models for Long-term Autonomy
Acshi Haggenmiller presents research on localization, mapping, planning and moving.
PhD Defense: Planning, Control, and Estimation for Diverse Multi-UAS Missions
Matthew Romano presents research on methods to command and control teams of Uncrewed Aerial Systems.
Shannon Danforth, also winner of Best RAL Paper Award at BioRob 2022, presents on preventing falls in older adults and people with lower-limb amputation through analyzing sit-to-stand motions and utilizing robotic exoskeletons and prostheses.
The Hidden Genius Project visits Michigan Robotics
Several students and research groups showed off their work in prosthetics, exoskeletons, legged robots and more during this outreach event.
CLAWS student team develops tools to help astronauts navigate space
Leia Stirling advises the Collaborative Lab for Advancing Work in Space, which participated in the NASA SUITS Challenge. The focus of this challenge was to develop an augmented reality system to help astronauts navigate the moon in the Artemis missions.
Inspired by ROB 101: Robotics for 11th graders
Take a look inside Isra Elshafei’s robotics class at The School at Marygrove, working with the School of Education.