Happening @ Michigan https://events.umich.edu/list/rss RSS Feed for Happening @ Michigan Events at the University of Michigan. Bioethics Discussion: Artificial Intelligence (September 29, 2020 7:00pm) https://events.umich.edu/event/58828 58828-14563719@events.umich.edu Event Begins: Tuesday, September 29, 2020 7:00pm
Location: Lurie Biomedical Engineering
Organized By: The Bioethics Discussion Group

A discussion on how we know machines know.

Here are a few readings to consider:
––Ethical Issues of Artificial Intelligence in Medicine
––Regulatory responses to medical machine learning
––Will artificial intelligence solve the human resource crisis in healthcare?
––Medical ethics considerations on artificial intelligence

For more information and/or to receive a copy of the readings visit http://belmont.bme.umich.edu/bioethics-discussion-group/discussions/047-artificial-intelligence/.

––

While people are still allowed on campus, discussions will be held on the front lawn of Lurie Biomedical Engineering building. Participants will be asked to enter the area via a “welcome desk” where there will be hand sanitizer, wipes, etc. Participants will be masked, at least 12 feet from one another, and speaking through megaphones with one another. In accordance with public health mandates and guidance, participation will be limited to 20 individuals who sign up to participate ahead of time.

Sign up here: https://belmont.bme.umich.edu/ask-your-questions-to-ponder/

––
One's intelligence might be artificially enhanced by the blog: https://belmont.bme.umich.edu/incidental-art/

]]>
Lecture / Discussion Tue, 25 Aug 2020 11:09:51 -0400 2020-09-29T19:00:00-04:00 2020-09-29T20:30:00-04:00 Lurie Biomedical Engineering The Bioethics Discussion Group Lecture / Discussion Artificial Intelligence
Friday Night AI (October 2, 2020 7:00pm) https://events.umich.edu/event/77588 77588-19885834@events.umich.edu Event Begins: Friday, October 2, 2020 7:00pm
Location: Off Campus Location
Organized By: Michigan Artificial Intelligence Laboratory

Invited speakers:
Profs. Ceren Budak (School of Information) and Rada Mihalcea (Michigan AI)
Organizer: Michigan AI Lab in collaboration with the Ann Arbor District Library
Moderator: Prof. Benjamin Kuipers, Michigan AI

Pre-registration required by Oct. 1.

The proliferation of misleading information in everyday access media outlets such as social media feeds, news blogs, and online newspapers has made it challenging to identify trustworthy news sources. Over the past few months, the amount of misinformation shared online has been exacerbated by the COVID-19 pandemic (which has sometime been also referred to as an infodemic) as well as by the ongoing political debates and the upcoming federal elections. Artificial Intelligence provides ways to identify misinformative content online, and to potentially curb its spread. Join us for a conversation with Michigan experts Prof. Ceren Budak and Prof. Rada Mihalcea, who will discuss how Artificial Intelligence can be used to address fake news and misinformation.

What are the ways that AI may be used to identify misinformation and fake news?
What are the challenges encountered when developing such AI systems?
What are the benefits and risks of using automated ways to fight misinformation?

About the panelists:
Ceren Budak is an Assistant Professor of Information, School of Information and Assistant Professor of Electrical Engineering and Computer Science, College of Engineering at the University of Michigan. Her research interests lie in the area of computational social science. She is particularly interested in the use of large scale data sets and computational techniques to study problems with policy, social and political implications.
Rada Mihalcea is a Professor of Computer Science and Engineering at the University of Michigan and the Director of the Michigan Artificial Intelligence Lab. Her research interests are in computational linguistics, with a focus on lexical semantics, multilingual natural language processing, and computational social sciences. Together with her research lab and collaborators, she has worked on the problem of automatic deception detection for more than ten years, addressing among others the detection of deception in language and multimodal streams, the identification of fake news, and identity deception. She is the recipient of a Presidential Early Career Award for Scientists and Engineers awarded by President Obama (2009) and a Fellow of the Association for Computing Machinery (2019).

]]>
Lecture / Discussion Tue, 22 Sep 2020 15:36:26 -0400 2020-10-02T19:00:00-04:00 2020-10-02T20:30:00-04:00 Off Campus Location Michigan Artificial Intelligence Laboratory Lecture / Discussion Friday Night AI
MDP Project Preview Night (October 6, 2020 5:00pm) https://events.umich.edu/event/78172 78172-19989036@events.umich.edu Event Begins: Tuesday, October 6, 2020 5:00pm
Location: Off Campus Location
Organized By: Multidisciplinary Design Program

Join us for a virtual project recruitment event, using the Career Fair Plus software.

- Talk to corporate sponsors and faculty PI’s about their projects
- Register ahead of time for interview slots similar to the engineering career fair
- Upload your resume and be prepared to ask questions
- Read the project descriptions on the MDP website before attending

]]>
Reception / Open House Mon, 05 Oct 2020 12:18:57 -0400 2020-10-06T17:00:00-04:00 2020-10-06T21:00:00-04:00 Off Campus Location Multidisciplinary Design Program Reception / Open House A Sponsor Mentor from the Human Rights First - Multidisciplinary Design Program project speaks with a prospective applicant at the Project Preview Night event in 2019.
TechArb Entrepreneurial Meetup (October 6, 2020 5:00pm) https://events.umich.edu/event/77699 77699-19901733@events.umich.edu Event Begins: Tuesday, October 6, 2020 5:00pm
Location: Off Campus Location
Organized By: Center for Entrepreneurship

During this virtual event, you'll hear from Edi Demaj from KodeLabs about building an international startup portfolio and network with other students while sharing your next big idea or finding student startups to join.

Networking can be difficult to do online, but we promise you'll listen to wonderful speakers and participate in interactive networking! You'll end the session having had fun meeting several new entrepreneurs and innovators.

]]>
Social / Informal Gathering Thu, 24 Sep 2020 15:47:33 -0400 2020-10-06T17:00:00-04:00 2020-10-06T18:30:00-04:00 Off Campus Location Center for Entrepreneurship Social / Informal Gathering Event Description
MDP Project Fair (October 7, 2020 12:00pm) https://events.umich.edu/event/78176 78176-19989039@events.umich.edu Event Begins: Wednesday, October 7, 2020 12:00pm
Location: Off Campus Location
Organized By: Multidisciplinary Design Program

Join us for a virtual project recruitment event, using the Career Fair Plus software.

- Talk to corporate sponsors and faculty PI’s about their projects
- Register ahead of time for interview slots similar to the engineering career fair
- Upload your resume and be prepared to ask questions
- Read the project descriptions on the MDP website before attending

]]>
Reception / Open House Mon, 05 Oct 2020 12:18:54 -0400 2020-10-07T12:00:00-04:00 2020-10-07T16:00:00-04:00 Off Campus Location Multidisciplinary Design Program Reception / Open House A Sponsor Mentor and a Faculty Mentor from the Northrop Grumman Solar Truss - Multidisciplinary Design Program project speak with a prospective applicant at the Project Preview Night event in 2019.
EER Seminar Series (Engineering Education Research) (October 7, 2020 3:30pm) https://events.umich.edu/event/77660 77660-19899719@events.umich.edu Event Begins: Wednesday, October 7, 2020 3:30pm
Location: Off Campus Location
Organized By: Engineering Education Research

Advancing diversity, inclusion, and equity has been a persistent challenge in engineering. Over the last 40 years, hundreds of papers and more than 25 national reports have been published focusing on broadening participation in STEM. Simultaneously, people throughout the U.S. have been working endlessly to solve this problem. Yet, we have seen only incremental progress, suggesting that there is a need to take a step back and re-examine what has been done, in terms of both research and practice. To support this effort, Dr. Lee’s research focuses on critically evaluating the research-to-practice cycle as it relates to broadening participation. In this seminar, he will discuss an ongoing project focused on the participation of Black Americans in engineering and computer science. The goals of this project are to advance our understanding of the disconnect between research and practice, to identify barriers to progress, and to set a national agenda for broadening the participation of Black Americans in engineering and computer science.

]]>
Lecture / Discussion Thu, 24 Sep 2020 09:41:43 -0400 2020-10-07T15:30:00-04:00 2020-10-07T16:30:00-04:00 Off Campus Location Engineering Education Research Lecture / Discussion Dr. Walter Lee
CFE TechLab Programs Info Session (October 8, 2020 4:30pm) https://events.umich.edu/event/77446 77446-19854031@events.umich.edu Event Begins: Thursday, October 8, 2020 4:30pm
Location: Off Campus Location
Organized By: Center for Entrepreneurship

This is your opportunity to ask TechLab staff and instructors anything and everything! During this info session, we’ll go deeper into program specifics and have you leave with a better understanding of how TechLab Climate Change and TechLab at Mcity can help you with your entrepreneurial career goals.

]]>
Presentation Fri, 18 Sep 2020 14:31:58 -0400 2020-10-08T16:30:00-04:00 2020-10-08T18:00:00-04:00 Off Campus Location Center for Entrepreneurship Presentation Students visiting IA Ventures in D.C.
2020 Virtual EER Prospective Student Open House (October 9, 2020 10:00am) https://events.umich.edu/event/77361 77361-19844064@events.umich.edu Event Begins: Friday, October 9, 2020 10:00am
Location: Off Campus Location
Organized By: Engineering Education Research

Students from all institutions are invited to attend. Participants will hear all about the program, meet the faculty and graduate students, learn about career opportunities as a UM graduate in this field and take a virtual tour of the beautiful University of Michigan campus.

Please note that applicants to the EER graduate program must have a Bachelor's and Master's degree in a traditional engineering discipline.

]]>
Lecture / Discussion Tue, 29 Sep 2020 14:53:47 -0400 2020-10-09T10:00:00-04:00 2020-10-09T16:30:00-04:00 Off Campus Location Engineering Education Research Lecture / Discussion EER Logo
Bioethics Discussion: Artificial Parts (October 13, 2020 5:00pm) https://events.umich.edu/event/58829 58829-14563720@events.umich.edu Event Begins: Tuesday, October 13, 2020 5:00pm
Location: Lurie Biomedical Engineering
Organized By: The Bioethics Discussion Group

A discussion on what is replaceable.

For the discussion, consider a few readings:
––Implant ethics
––Neuro-Prosthetics, the Extended Mind, and Respect for Persons with Disability
––Why Not Artificial Wombs?
––Going Out on a Limb: Prosthetics, Normalcy and Disputing the Therapy/Enhancement Distinction

For more information and/or to receive a copy of the readings visit http://belmont.bme.umich.edu/bioethics-discussion-group/discussions/048-artificial-parts/.

––

While people are still allowed on campus, discussions will be held on the front lawn of Lurie Biomedical Engineering building. Participants will be asked to enter the area via a “welcome desk” where there will be hand sanitizer, wipes, etc. Participants will be masked, at least 12 feet from one another, and speaking through megaphones with one another. In accordance with public health mandates and guidance, participation will be limited to 20 individuals who sign up to participate ahead of time.

Sign up here: https://belmont.bme.umich.edu/ask-your-questions-to-ponder/

––
Part way between "the real" and "the artificial", "the blog": https://belmont.bme.umich.edu/incidental-art/

]]>
Lecture / Discussion Mon, 12 Oct 2020 20:42:47 -0400 2020-10-13T17:00:00-04:00 2020-10-13T18:30:00-04:00 Lurie Biomedical Engineering The Bioethics Discussion Group Lecture / Discussion Artificial Parts
(Re)Engaging the Role of Diversity, Equity, and Inclusion in Engineering Graduate Education (October 28, 2020 3:30pm) https://events.umich.edu/event/78529 78529-20058230@events.umich.edu Event Begins: Wednesday, October 28, 2020 3:30pm
Location: Off Campus Location
Organized By: Engineering Education Research

Increasingly, engineering graduate programs have emphasized the need to train individuals who are capable of working in diverse teams so they are better able to address complex problems in a global society. Yet, discourse related to diversity, equity, and inclusion (DEI) in engineering is often focused on recruiting and retaining students who are racially minoritized and/or women in the field. Less attention is given to what students learn about DEI during their graduate training. Drawing from findings across multiple research projects, this talk will explore what graduate students learn about the role of DEI in engineering and the implications these lessons have for racially minoritized students’ retention, success, and career pathways.

BIOGRAPHICAL SKETCH: Dr. Rosemary (Rosie) Perez is an Associate Professor in the Center for the Study of Higher and Postsecondary Education at the University of Michigan. She earned her B.S. in biological sciences and psychology at Carnegie Mellon University, her M.Ed. in higher education and student affairs at The University of Vermont, and her Ph.D. in higher education from University of Michigan. Dr. Perez’s scholarship has three interrelated lines of inquiry and explores: (a) how people make meaning of collegiate experiences; (b) diverse learning environments and intercultural development; and (c) the professional socialization of graduate students and new practitioners. Across projects, Dr. Perez explores the tensions between structure and agency, and how power, privilege, and oppression affect individuals and groups within higher education. Her research has been funded by the National Science Foundation, Spencer Foundation, Susan Thompson Buffett Foundation, and ACPA-College Student Educators International.

]]>
Lecture / Discussion Wed, 21 Oct 2020 12:58:11 -0400 2020-10-28T15:30:00-04:00 2020-10-28T16:30:00-04:00 Off Campus Location Engineering Education Research Lecture / Discussion Dr. Rosemary Perez
Dignifying the Disinherited: The Case for Pro-Black Engineering Education Research (November 18, 2020 3:30pm) https://events.umich.edu/event/79149 79149-20217705@events.umich.edu Event Begins: Wednesday, November 18, 2020 3:30pm
Location: Off Campus Location
Organized By: Engineering Education Research

Research is essential to the infrastructure of education and plays a prominent role in driving curriculum, policy, and professional practice. Therefore, engineering education research (EER) is critical to driving the impetus and approach to racial equity within engineering education and practice. This presentation will spotlight how anti-Blackness is embedded in EER practices and delineate its roots in America’s systemic racism. Centering the experience of Black people within the engineering education community helps reframe the problem of racial/ethnic exclusion, while generating a new way forward through pro-Black EER (PEER). PEER uses critical methodologies, frameworks, and intentional citation practices to assert the genius of Black people.

]]>
Lecture / Discussion Tue, 03 Nov 2020 12:47:59 -0500 2020-11-18T15:30:00-05:00 2020-11-18T16:30:00-05:00 Off Campus Location Engineering Education Research Lecture / Discussion Dr. James Holly, Jr.
Collection and Analysis of Driving Videos based on Traffic Participants (December 3, 2020 4:00pm) https://events.umich.edu/event/79591 79591-20428440@events.umich.edu Event Begins: Thursday, December 3, 2020 4:00pm
Location: Off Campus Location
Organized By: Michigan Robotics

Autonomous vehicle (AV) prototypes have been deployed in increasingly varied environments in recent years. An AV must be able to reliably detect and predict the future motion of traffic participants to maintain safe operation based on data collected from high-quality onboard sensors. Sensors such as camera and LiDAR generate high-bandwidth data that requires substantial computational and memory resources. To address these AV challenges, this thesis investigates three related problems: 1) What will the observed traffic participants do? 2) Is an anomalous traffic event likely to happen in near future? and 3) How should we collect fleet-wide high-bandwidth data based on 1) and 2) over the long-term?

The first problem is addressed with future traffic trajectory and pedestrian behavior prediction.We propose a future object localization (FOL) method for trajectory prediction in first person videos (FPV). FOL encodes heterogeneous observations including bounding boxes, optical flow features and ego camera motions with multi-stream recurrent neural networks (RNN) to predict future trajectories. We then introduce BiTraP, a goal-conditioned bidirectional multi-modal trajectory prediction method. BiTraP estimates multi-modal trajectories and uses novel bi-directional decoder and loss to improve longer-term trajectory prediction accuracy. We show that different choices of non-parametric versus parametric target models directly influence predicted multi-modal trajectory distributions. Experiments with two FPV and six bird's-eye view (BEV) datasets show the effectiveness of our methods compared to state-of-the-art. We define pedestrian behavior prediction as a combination of action and intent. We hypothesize that current and future actions are strong intent priors and propose a multi-task leaning RNN encoder-decoder network to detect and predict future pedestrian actions and street crossing intent. Experimental results show that one task helps the other so they together achieve state-of-the-art performance on published datasets.

To identify likely traffic anomaly events, we propose to predict locations of traffic participants over a near-term future horizon and monitor accuracy and consistency of these predictions as evidence of an anomaly. Inconsistent predictions tend to indicate an anomaly has or is about to occur. A supervised video action recognition method can then be applied to classify detected anomalies. We introduce a spatial-temporal area under curve (STAUC) metric as a supplement to the existing area under curve (AUC) evaluation and show it captures how well a model detects both temporal and spatial locations of anomalous events. Experimental results show the proposed method and consistency-based anomaly score are more robust to moving cameras than image generation based methods; our method achieves state-of-the-art performance over AUC and STAUC metrics.

Video anomaly detection (VAD) and action recognition support event-of-interest (EOI) distinction from normal driving data. We introduce a Smart Black Box (SBB), an intelligent event data recorder, to prioritize EOI data in long-term driving. The SBB compresses high-bandwidth data based on EOI potential and on-board storage limits. The SBB is designed to prioritize newer and anomalous driving data and discard older and normal data. An optimal compression factor is selected based on the trade-off between data value and storage cost.Experiments in a traffic simulator and with real-world datasets show the efficiency and effectiveness of using a SBB to collect high-quality videos in long-term driving.

]]>
Livestream / Virtual Mon, 30 Nov 2020 11:27:05 -0500 2020-12-03T16:00:00-05:00 2020-12-03T18:00:00-05:00 Off Campus Location Michigan Robotics Livestream / Virtual car with bounding box
Bioethics Discussion: Annihilation (December 8, 2020 7:00pm) https://events.umich.edu/event/58833 58833-14563725@events.umich.edu Event Begins: Tuesday, December 8, 2020 7:00pm
Location: Lurie Biomedical Engineering
Organized By: The Bioethics Discussion Group

A discussion on our obliteration.

[Video-conference link: https://umich.zoom.us/j/94651294615]

A few readings to consider before oblivion:
–– Bioethics and the Metaphysics of Death
––The Ontological Representation of Death: A Scale to Measure the Idea of Annihilation Versus Passage
––The Nonidentity Problem and Bioethics: A Natural Law Perspective
––Controversies in the Determination of Death: A White Paper of the President’s Council on Bioethics

For more information and/or to receive a copy of the readings visit http://belmont.bme.umich.edu/bioethics-discussion-group/discussions/052-annihilation/.

––
When the server hosting this blog is turned off, where does the website go: https://belmont.bme.umich.edu/incidental-art/?

]]>
Lecture / Discussion Tue, 08 Dec 2020 15:46:52 -0500 2020-12-08T19:00:00-05:00 2020-12-08T20:30:00-05:00 Lurie Biomedical Engineering The Bioethics Discussion Group Lecture / Discussion Annihilation
Data-Driven Methods for Geometric Systems (December 14, 2020 12:30pm) https://events.umich.edu/event/79750 79750-20483939@events.umich.edu Event Begins: Monday, December 14, 2020 12:30pm
Location: Off Campus Location
Organized By: Michigan Robotics

Recently, robots have played an increasingly helpful role in navigation, mapping, remote manipulation, and many other dynamic applications. As capabilities continue to advance, robots with many joints offer the potential to execute more nuanced, sophisticated tasks than simpler mechanisms. However, the curse of dimensionality can place prohibitive costs in time and resources in order to control and refine such behaviors. In this work, we investigated the role of system geometry in addressing these challenges. Geometric mechanics offers a framework to generalize intuitive features, like friction and inertia, into broad categories of robots that experience the same functional forms relating momentum, internal shape motions, and body motions. We focused on friction-dominated robots, where we could see that the vanishing role of momentum reduces the dynamics from a second order to a first order system. Subsequent architectural simplifications in behavior modeling, planning, and control resulted in robots that were capable of rapidly self-modeling and optimizing useful behaviors. We demonstrated on a simulated robotic snake that during joint failure, the system was able to adapt more quickly when it was equipped with more motorized joints. In this case, dimensionality was an asset, rather than a liability. Additionally, we demonstrated that the methods use no prior knowledge about system kinematics by building a robot made of tree branches. This system was able to optimize a library of primitives for navigation, training on less than 12 minutes of experimental data. Finally we showed that these methods can extend to both soft robots and systems with momentum. This defense will cover our findings concerning the practical applications of data-driven geometric mechanics.

]]>
Livestream / Virtual Mon, 07 Dec 2020 13:27:15 -0500 2020-12-14T12:30:00-05:00 2020-12-14T13:30:00-05:00 Off Campus Location Michigan Robotics Livestream / Virtual tree branch robot
Metalevel Motion Planning for Unmanned Aircraft Systems: Metrics Definition and Algorithm Selection (January 26, 2021 2:00pm) https://events.umich.edu/event/81051 81051-20838703@events.umich.edu Event Begins: Tuesday, January 26, 2021 2:00pm
Location: Off Campus Location
Organized By: Michigan Robotics

A diverse suite of manned and unmanned aircraft will occupy future urban airspace. Flight plans must accommodate specific aircraft characteristics including physical volume with safety zone clearance, landing/takeoff procedures, kinodynamics, and a wide range of flight environments. No single motion planner is applicable across all possible aircraft configurations and operating conditions. This dissertation proposes the first motion planning algorithm selection capability with application to small Unmanned Aircraft System (UAS) multicopters operating in and over a complex urban landscape.

Fail-safe studies are presented to improve on contemporary "fly-home" or automatic landing protocols. Three alternative data-driven fail-safe protocols are presented for multicopter UAS urban flight, focusing on building rooftops as safe urban landing sites. In the first fail-safe strategy, the multicopter identifies, generates, and follows a flight plan to the closest available rooftop suitable for landing. In the second supervisory fail-safe strategy, the multicopter examines rooftops en route to a planned landing site, diverting to a closer, clear landing site when possible. In the third fail-safe strategy, the multicopter cannot preplan a safe landing site due to missing landing site data. In this case, the multicopter executes a coverage path to explore the area and evaluate overflown rooftops to find a safe landing site. These three fail-safe algorithms integrate map generation, flight planning, and area coverage capabilities.

The motion planning algorithm selection problem (ASP) requires qualitative and quantitative metrics to inform the ASP of user/agent, algorithm, and configuration space preferences and constraints. Urban flight map-based, path-based, and software-based cost metrics are defined to provide insights into the properties of the urban canyon needed to construct safe and efficient flight plans. Map-based metrics describe the operating environment by constructing a collection of GPS/Lidar navigation performance, population density, and obstacle risk exposure metric maps. Path-based metrics account for a vehicle's energy consumption and distance traveled. Software-based metrics measure memory consumption and execution time of an algorithm. Metric maps were analyzed in-depth with path-based and software-based metrics utilized in Monte Carlo and ASP studies.

An algorithm portfolio consisting of geometric (Point-to-Point: PTP), graph-based (A* variants), and sampling-based (BIT* variants) motion planners were considered in this work. Path cost, execution time, and success rate benchmarks were investigated using Monte Carlo problem instances with A* "plus" producing the lowest cost paths, PTP having the fastest executions, and A* "dist" having the best overall success rates. The BIT* variant paths typically had higher cost but their success rate increased relative to altitude. The problem instances and metric maps informed two new machine learning solutions for urban small UAS motion planning ASP. Rule-based decision trees were simple to construct but unable to capture both complex cost metrics and algorithm properties. The investigated neural network-based ASP formulations produced promising results, with a hybrid two-stage selection scheme having the best algorithm selection accuracy, laying the seeds for future work.

The most significant innovation of this dissertation is motion planning ASP for UAS. Non-traditional open-source databases also advance the field of data-driven flight planning, contributing to fail-safe UAS operations as well as ASP. Path planning algorithms integrated a new suite of diverse cost metrics accompanied by a novel multi-objective admissible heuristic function. Neural network and decision tree ASP options were presented and evaluated as a first-case practical approach to solving the motion planning ASP for small UAS urban flight.

]]>
Livestream / Virtual Thu, 21 Jan 2021 11:31:54 -0500 2021-01-26T14:00:00-05:00 2021-01-26T15:00:00-05:00 Off Campus Location Michigan Robotics Livestream / Virtual chart
Supporting the Integration of Numerical Computation in Physics Education (January 27, 2021 3:30pm) https://events.umich.edu/event/80602 80602-20761740@events.umich.edu Event Begins: Wednesday, January 27, 2021 3:30pm
Location: Off Campus Location
Organized By: Engineering Education Research

Computation has revolutionized how modern science is done. Modern scientists use computational techniques to reduce mountains of data, to simulate impossible experiments, and to develop intuition about the behavior of complex systems. Much of the research completed by modern scientists would be impossible without the use of computation. And yet, while computation is a crucial tool of practicing scientists, most modern science curricula do not reflect its importance and utility. In this talk, I will discuss the urgent need to construct such curricula and present research that investigates the challenges at a variety of scales from the large (institutional structures) to the small (student understanding of a concept). I will discuss how the results of this research can be leveraged to facilitate the computational revolution in science education. This research will help us understand and develop institutional incentives, effective teaching practices, evidence-based course activities, and valid assessment tools. This work has been supported by Michigan State University’s CREATE for STEM Institute, the National Science Foundation, the Norwegian Agency for Quality Assurance in Education (NOKUT), the Norwegian Research Council, and the Thon Foundation.

]]>
Lecture / Discussion Mon, 11 Jan 2021 12:45:09 -0500 2021-01-27T15:30:00-05:00 2021-01-27T16:30:00-05:00 Off Campus Location Engineering Education Research Lecture / Discussion Danny Caballero
An Honest Conversation: Diversity and Inclusion in Engineering (February 17, 2021 3:30pm) https://events.umich.edu/event/81702 81702-20943454@events.umich.edu Event Begins: Wednesday, February 17, 2021 3:30pm
Location: Off Campus Location
Organized By: Engineering Education Research

In 2020, we witnessed several examples of social injustice and social unrest. As human beings and engineers, we must decide how we want to respond to what happened and how we want to move forward. Calls to improve our approach to diversity, equity, and inclusion (DEI) have coincided with the call to update our overall engineering curriculum. Over the years, several initiatives have been launched to address such issues, which primarily attempt to address perceived inadequacies in underrepresented students. However, scarce efforts have been developed to address the engineering culture that has limited the full participation of women and people of color in engineering. Furthermore, few of us in engineering have the knowledge, skills, or ability to productively engage with issues leading to the marginalization and social unrest. Rarely do we dare to apply our problem-solving or critical thinking approaches to how to educate or improving DEI. As a result, the goal of this talk is to provide engineers with language to have an honest conversation about our individual and collective response to the inequity in engineering and realign our actions to improve engineering education. This impactful workshop will provide definitions and practical examples of key DEI concepts in engineering based on holistic interdisciplinary research.

]]>
Lecture / Discussion Thu, 04 Feb 2021 08:44:52 -0500 2021-02-17T15:30:00-05:00 2021-02-17T16:30:00-05:00 Off Campus Location Engineering Education Research Lecture / Discussion Dr. Kelly Cross
Motivation and Identity as Signals of Systemic Problems in Engineering Education (March 10, 2021 3:30pm) https://events.umich.edu/event/82513 82513-21114065@events.umich.edu Event Begins: Wednesday, March 10, 2021 3:30pm
Location: Off Campus Location
Organized By: Engineering Education Research

There is a well-documented history of systemic engineering education problems ranging from a persistently chilly climate to a burgeoning mental health crisis. Outcomes of these problems include but are not limited to increased attrition, decreased learning, and reduced engineering innovation resulting from a homogenous engineering population. While these measures provide concrete evidence of systemic problems, they do not provide clear targets for change or an early warning system of how systemic problems influence students before crucial decision points.

To address the limitations of existing engineering education outcome measures, measures of how students internalize engineering experiences are needed. Students' motivations for engineering tasks and identifications as engineers can fill this gap as they are contextually responsive and connected to educational outcomes such as deep learning, student retention, and task persistence. Additionally, students' educational experiences directly influence their motivations and identities.

Informed by specific theories of motivation and identity (future time perspective and engineering role identity, respectively), this presentation describes how students' motivations and identities are shaped by their engineering education experiences and shape engineering education cultures. Specifically, I will discuss the homogenization of undergraduates' motivations and identities; the connections between motivation and identity and experiences of discrimination and bias; and the identity and motivationally undermining experiences of engineering graduate students. I will conclude by discussing actionable steps to shift engineering education defaults to foster students' motivations and identities.

Biographical Sketch: Dr. Adam Kirn is an Associate Professor of Engineering Education in the Department of Electrical and Biomedical Engineering at the University of Nevada, Reno. His research focuses on the ways students' motivations and identities shape and are shaped by their engineering education experiences. The results of this work seek to implement evidence-based practices to create educational defaults that foster student success and thriving. Adam has a B.S. in Biomedical Engineering from Rose-Hulman Institute of Technology, an M.S. in Bioengineer, and a Ph.D. in Engineering and Science Education from Clemson University.

]]>
Lecture / Discussion Thu, 25 Feb 2021 09:26:14 -0500 2021-03-10T15:30:00-05:00 2021-03-10T16:30:00-05:00 Off Campus Location Engineering Education Research Lecture / Discussion Dr. Adam Kirn
Community Cultural Wealth, Program Evaluation, and ASEE CDEI, Oh My! (March 24, 2021 3:30pm) https://events.umich.edu/event/83003 83003-21235293@events.umich.edu Event Begins: Wednesday, March 24, 2021 3:30pm
Location: Off Campus Location
Organized By: Engineering Education Research

As a sociologist who has been working in STEM and Engineering Education for 18 years, and who isn’t on the tenure track, Liz will share a little bit about a few different areas (Research, Evaluation, and Service) that she has focused on in her career.  Assets-based frameworks for understanding student experience are receiving more and more visibility these days and Liz’s work has used critical race theory Community Cultural Wealth (Samuelson & Litzler, JEE 2016) to understand the ways minoritized engineering undergraduates deployed their cultural assets to persist in engineering. She is also now working on further Community Cultural Wealth research with her colleagues on the PNW-LSAMP project.  She will also talk about using her social science research skills to conduct high quality program evaluation of projects focused on improving DEI in STEM.  Finally, she’ll share about the work of the ASEE Commission on Diversity, Equity, and Inclusion, of which she is the current chair. CDEI is a great resource for the community and also a wonderful opportunity to develop new connections with colleagues while providing important service to the engineering education field.

]]>
Lecture / Discussion Fri, 12 Mar 2021 14:42:39 -0500 2021-03-24T15:30:00-04:00 2021-03-24T16:30:00-04:00 Off Campus Location Engineering Education Research Lecture / Discussion Dr. Elizabeth Litzler
Robotic Manipulation under Transparency and Translucency from Light-field Sensing (March 31, 2021 1:00pm) https://events.umich.edu/event/83269 83269-21328379@events.umich.edu Event Begins: Wednesday, March 31, 2021 1:00pm
Location: Off Campus Location
Organized By: Michigan Robotics

From frosted windows to plastic containers to refractive fluids, transparency and translucency are prevalent in human environments. The material properties of translucent objects challenge many of our assumptions in robotic perception. For example, the most common RGB-D sensors in robotic applications require the sensing of an infrared structured pattern from a Lambertian reflectance of surfaces. As such, transparent and translucent objects often remain invisible to robot perception. Thus, introducing methods that would enable robots to correctly perceive and then interact with the environment would be highly beneficial. Light-field (or plenoptic) cameras, for instance, which carry light direction and intensity, make it possible to perceive visual clues on object surfaces from reflection and refraction. In this dissertation, we explore the inference of transparent and translucent objects from plenoptic observations for robotic perception and manipulation. We propose a novel plenoptic descriptor, Depth Likelihood Volume (DLV), that incorporates multi-view plenoptic observations to represent depth. The depth of a pixel is then represented as a distribution rather than a single value. Through the DLV distribution, we can infer the layered translucency structure of the scene with transparent and translucent objects for robot manipulation tasks. Building on the DLV, we present the Plenoptic Monte Carlo Localization algorithm, PMCL, as a generative method to infer 6-DoF poses of objects in settings with translucency. PMCL is able to localize both isolated transparent objects and opaque objects behind translucent objects using a previously computed DLV. We evaluate PMCL by comparing x estimated poses against ground-truth poses and demonstrate the use of these pose estimates for object pick and place tasks. The uncertainty induced by transparency and translucency for pose estimation increases greatly as scenes become more cluttered. However, robot grasping does not necessarily require estimation of 6-DoF object poses. Given multi-view plenoptic observations, we propose GlassLoc to localize feasible grasp poses over a pile of transparent objects. In GlassLoc, a convolutional neural network is introduced to learn DLV features for classifying grasp poses with grasping confidence. GlassLoc also suppresses the reflectance by checking pixel consistency over multi-view plenoptic observations, which leads to more stable DLV representation. We evaluate GlassLoc in the context of a pick-and-place task for transparent tableware in a cluttered tabletop environment. We further observe that the transparent and translucent objects will generate distinguishable features in the light-field epipolar image plane, which provides information about the object location. With this insight, we propose Light-field Inference of Transparency, LIT, as a two-stage generative-discriminative refractive object localization approach. In the discriminative stage, LIT uses convolutional neural networks to learn reflection and distortion features from photorealistic-rendered light-field images. The learned features guide generative object location inference through local depth estimation and particle optimization. With the LIT pipeline, we also create the light-field dataset for the task of transparent objects recognition, segmentation, and pose estimation. We compare LIT with three state-of-the-art pose estimators to show our efficacy in the transparent object localization task. We also perform a robot demonstration by picking champagne cups up from a textureless table and building a champagne tower using the LIT pipeline.

]]>
Livestream / Virtual Wed, 24 Mar 2021 10:48:06 -0400 2021-03-31T13:00:00-04:00 2021-03-31T15:00:00-04:00 Off Campus Location Michigan Robotics Livestream / Virtual Robot holds a champagne glass
Assessing the Progress of Diversity, Equity, and Inclusion Initiatives in the College of Engineering: Student Perceptions of the Climate at the University of Michigan (April 14, 2021 3:30pm) https://events.umich.edu/event/83292 83292-21367800@events.umich.edu Event Begins: Wednesday, April 14, 2021 3:30pm
Location: Off Campus Location
Organized By: Engineering Education Research

The University of Michigan College of Engineering is nearing the end of its five-year strategic plan to improve the climate on campus with regards to Diversity, Equity, and Inclusion (DEI). As part of that plan, focus groups were held with students across every department and program in the College of Engineering, to gather qualitative data that can serve as metrics to consider how the DEI strategic plan is progressing. Over thirty focus groups were held with more than 220 undergraduate and graduate students across the College in the 2019-2020 and 2020-2021 academic years. Students were asked about various elements of their perception of DEI, including factors that impacted their sense of inclusion, if they had been treated differently based on their identity, and their perceptions of the College and their department with regards to diversity. The data collected in these focus groups illuminates the nuance and complexity of the engineering student experience, and how that experience, and their perceptions of DEI in the College, can vary based on their identities and home department.

]]>
Lecture / Discussion Mon, 29 Mar 2021 10:00:26 -0400 2021-04-14T15:30:00-04:00 2021-04-14T16:30:00-04:00 Off Campus Location Engineering Education Research Lecture / Discussion Dr. Laura Hirshfield
Implantable Neural Interfaces for Direct Control of Hand Prostheses (April 27, 2021 10:00am) https://events.umich.edu/event/83793 83793-21530358@events.umich.edu Event Begins: Tuesday, April 27, 2021 10:00am
Location: Off Campus Location
Organized By: Michigan Robotics

State of the art robotic hands can mimic many functions of the human hand. These devices are a capable of actuating individual finger and multi-joint movements while providing adequate gripping force for daily activities. However, for patients with spinal cord injuries or amputations, there are few options to control these functions seamlessly or intuitively. A common barrier to restoring hand function to both populations is a lack of high-fidelity control signals. Non-invasive electrophysiological techniques record global summations of activity and lack the spatial or temporal resolution to extract or “decode” precise movement commands. The ability to decode finger movements from the motor system would allow patients to directly control hand functions and provide intuitive and scalable prosthetic solutions. This thesis investigates the capabilities of implantable devices to provide finger-specific commands for prosthetic hands. We adapt existing reasoning algorithms to two different sensing technologies.

The first is intracortical electrode arrays implanted into primary motor cortex of two non-human primates. Both subjects controlled a virtual hand with a regression algorithm that decoded their brain activity into finger kinematics. Performance was evaluated with a single degree of freedom target matching task. A state-of-the-art re-calibration approach improved performance, measured as bit rate, by an average of 33.1%. Notably, decoding performance was not dependent on subjects moving their intact hands. In future research, this approach can improve grasp precision for patients with spinal cord injuries.

The second sensing technology is intramuscular electrodes implanted into residual muscles and Regenerative Peripheral Nerve Interfaces of two patients with transradial amputations. Both participants used a high-speed pattern recognition system to switch between 10 individual finger and wrist postures in a virtual environment with an average success rate of 94.7% and a trial latency of 255 ms. When the set was reduced to five grasp postures, average metrics improved to 100% success and 135 ms latency. These results are a significant improvement over state-of-the-art systems that use surface electromyography as inputs. Furthermore, grasp performance remained reliable across arm positions and both participants used this controller to complete a functional assessment with robotic prostheses.
For a more dexterous solution, we combined the high-speed pattern recognition system with a regression algorithm that enabled simultaneous position control of both the index finger and middle-ring-small finger group. Both patients used this system to complete a virtual two degree of freedom target matching task with throughputs of 1.79 and 1.15 bits per second each. The controllers in this study used only four and five differentiated inputs, which can feasibly be processed with portable or implantable hardware.

These results demonstrate that implantable systems can provide patients with fluid and precise control of hand prostheses, eliminating the need to use movement substitutions and triggers to cycle through grip modes. However, clinically translatable implantable electronics need to be developed to realize the potential of these sensing and reasoning approaches. Further advancement of this technology will likely increase the utility and demand of robotic prostheses.

]]>
Livestream / Virtual Mon, 19 Apr 2021 13:55:31 -0400 2021-04-27T10:00:00-04:00 2021-04-27T12:00:00-04:00 Off Campus Location Michigan Robotics Livestream / Virtual prosthetic hand stacks block
Interpretable and Realtime Predictions of Social Interactions for Autonomous Vehicles (April 29, 2021 3:00pm) https://events.umich.edu/event/83792 83792-21530357@events.umich.edu Event Begins: Thursday, April 29, 2021 3:00pm
Location: Off Campus Location
Organized By: Michigan Robotics

Autonomous vehicles present an opportunity to transform transportation. The benefits range from increased access to mobility and time freed from driving, to greater safety due to automation. These robots are powered by the coordination of various systems to perceive the world and effect motion control. Crucially, the autonomous vehicle operates in an open environment alongside fellow road users with whom it will interact regularly. Predictions of fellow road users' intents and future motion guide these interactions and specify a large part of the autonomous vehicle's behavior. Spurred by advances in deep learning, recent prediction methods have increasingly begun to consider how interactions affect future motion in ever more varied environments. The corresponding gains in accuracy translate to more anticipatory and less reactive autonomous vehicle behavior. One drawback is an increase in complexity, which can lead to less interpretable predictions and behavior. Achieving realtime performance and handling missing data caused by adverse sensing conditions present additional challenges.

To support autonomous vehicle behavior that is transparent and predictable, this thesis develops interpretable prediction methods. Model-based approaches provide the vehicle for building interpretable predictions, and novel inference procedures are developed to generate the predictions in realtime. Adopting a probabilistic framework enables natural handling of missing data and affords the flexibility to model interactions in varied environments beyond those described by existing interpretable methods. Experiments on real highway traffic and urban data demonstrate the developed methods' effectiveness.

]]>
Livestream / Virtual Mon, 19 Apr 2021 10:09:07 -0400 2021-04-29T15:00:00-04:00 2021-04-29T17:00:00-04:00 Off Campus Location Michigan Robotics Livestream / Virtual autonomous car merging
Real-time Human Workload Estimation and Its Application in Adaptive Haptic Shared Control (May 25, 2021 1:30pm) https://events.umich.edu/event/84054 84054-21619770@events.umich.edu Event Begins: Tuesday, May 25, 2021 1:30pm
Location: Off Campus Location
Organized By: Michigan Robotics

Automated vehicles (AVs) are promising to have the potential to reduce driving-related injuries and deaths. However, autonomous driving technology is currently limited in its scope and reliability, giving rise to the semi-autonomous driving model, where the autonomy and the human share the control of the vehicle. Workload, despite being an important human factor, has not yet been considered when designing adaptive shared control. Recently, researchers have started to apply machine learning techniques to classify mental workload into different levels. However, most of these studies have adopted either a single-model-single-feature approach or a single-model-all-features approach. However different machine learning models are suitable for different features, how to leverage different models for different features is critical.

To address these shortcomings and research gaps, the goals of this dissertation were to (1) examine whether and to what extent haptic shared control performance can be improved by incorporating operators' workload; (2) develop a computational model for workload estimation, and the model should be able to leverage different machine learning models that work best for different features; and (3) investigate the generalizability of the workload estimation model. To address these research goals, this dissertation was composed of four research phases with two pilot studies and four human subject experiments.

(1) Collaborating with Yifan Weng, Dr. Tulga Ersal, and Prof. Jeffrey Stein from the Department of Mechanical Engineering at the University of Michigan, we developed a teleoperated dual-task shared control simulation platform where the human shared control of a ground vehicle with autonomy while performing a surveillance task simultaneously. In addition, we developed a real-time eye-tracking system based on Tobii Pro Glasses 2 to measure the human gaze points in a world frame and pupil sizes.

(2) We proposed a workload-adaptive haptic shared control scheme together with our collaborators. We conducted two human subject experiments during this phase. The results indicated that the proposed workload-adaptive haptic shared control scheme can reduce human workload, increase human trust in the system, increase driving performance, and reduce human effort without sacrificing surveillance task performance.

(3) We proposed a Bayesian inference model for workload estimation that can leverage the different machine learning models that work best for different features. Specifically, we used support-vector machines (SVMs) for pupil size change, the Hidden Markov Model (HMM) for gaze trajectory, SVMs for fixation feature, and Gaussian Mixture Models (GMMs) for fixation trajectory. The empirical results indicated that our proposed model achieved a 0.82 F1 score for workload imposed by varying surveillance task urgency.

(4) We investigated the generalizability of our proposed Bayesian inference model for workload estimation by conducting two human subject experiments with 24 participants and using different factors to impose human workload, i.e., obstacle headway and driving speed. The results indicated that our proposed model achieved a 0.70 F1 score for the workload imposed by obstacle avoidance and the personalized version of our proposed model can distinguish the workload imposed by different driving speeds under high surveillance task urgency.

]]>
Lecture / Discussion Mon, 17 May 2021 10:55:54 -0400 2021-05-25T13:30:00-04:00 2021-05-25T15:00:00-04:00 Off Campus Location Michigan Robotics Lecture / Discussion simulating driving while multitasking
EECS Juneteenth Celebration (June 18, 2021 1:00pm) https://events.umich.edu/event/84060 84060-21619781@events.umich.edu Event Begins: Friday, June 18, 2021 1:00pm
Location: Off Campus Location
Organized By: Electrical and Computer Engineering

https://umich.zoom.us/j/99331130203
Passcode: 719944

EECS invites you to our second Juneteenth celebration on Friday, June 18, from 1:00 – 2:00 p.m. ET.

Juneteenth is the oldest nationally celebrated commemoration of the ending of slavery in the United States. It celebrates African American freedom and achievement. Its goal is to promote and cultivate knowledge and appreciation of African American history and culture while encouraging continuous self-development and respect for all cultures. More info >

Our Juneteenth celebration will include:

-Performance of the Black National Anthem
-Reading of the Emancipation Proclamation
-Recognition of Dr. Willie Hobbs Moore, the first Black woman at Michigan to earn a BS and MS in Electrical Engineering (‘58 and ‘61), and the first Black woman in the country to earn a PhD in physics.
-Panel discussion on the representation of black students in STEM
All are welcome to attend to celebrate and learn!

If you have any problems accessing the event, please email eecs-comm@umich.edu.

Following the event, there will be an open Zoom meeting for additional questions and discussion.

https://umich.zoom.us/j/95094857220
Passcode: 010976

]]>
Livestream / Virtual Mon, 17 May 2021 12:08:50 -0400 2021-06-18T13:00:00-04:00 2021-06-18T14:00:00-04:00 Off Campus Location Electrical and Computer Engineering Livestream / Virtual
Optimal Task-Invariant Energetic Control for Powered Exoskeletons (June 22, 2021 8:00am) https://events.umich.edu/event/84328 84328-21623343@events.umich.edu Event Begins: Tuesday, June 22, 2021 8:00am
Location: Off Campus Location
Organized By: Michigan Robotics

Powered exoskeletons have been developed to serve as rehabilitation devices and provide gait assistance to human users. It enhances a healthy person's abilities and supports a physically challenged person's daily life by providing powered hip, knee, and/or ankle motions with different control designs. State-of-art powered exoskeletons use trajectory-based, kinematic control methods for specific tasks. This type of control is appropriate for paraplegia, where the exoskeleton provides complete assistance. However, it overly constrains the volitional motion of people with remnant voluntary ability, e.g., stroke patients. In contrast, trajectory-free control methods are now being developed to provide task-invariant partial assistance for practicing/relearning leg motions or performing a continuum of activities in varying environments. Robotics systems like powered exoskeletons can be represented in the format of the Euler-Lagrange equation or the equivalent class, the port-controlled Hamiltonian equations. We use energy shaping methods for task-invariance by altering the human-exoskeleton system's dynamic characteristics via the Euler-Lagrange equations or the more general port-controlled Hamiltonian equations. Under satisfaction of the matching conditions, which is a set of nonlinear partial differential equations (PDEs), the open-loop system's mass/inertia matrix and gravitational vector can be mapped to the desired closed-loop system dynamics.

Energy shaping has broad applicability but has been limited in part by its computational and analytical complexity. Changes to the system dynamics must follow the solution to the matching conditions, while finding a solution to the matching conditions itself is challenging. The corresponding controller for kinetic energy shaping requires complicated calculations of the mass/inertia matrix inverse, where the computational cost for implementation increases tremendously. Potential energy shaping avoids the matrix inversion by altering only the potential energy but prevents shaping on the inertia terms. Such challenges require a new total energy shaping (TES) framework that provides flexible shaping structures with more freedom to change closed-loop dynamics over the potential energy shaping method. The corresponding control law with shaped structure also requires efforts to optimally achieve the powered exoskeletons' target control strategy for different locomotor tasks. Therefore, the specific aims of this project are: 1) construct a TES framework for powered exoskeletons with flexible shaping structure, 2) apply the framework to powered exoskeletons for optimal assistance across varying locomotor tasks, and 3) perform experiments on multiple human subjects to demonstrate the possible clinical benefits of optimal TES framework with powered lower-limb exoskeletons.

This work is significant to the viability and performance of energy shaping methods to nonlinear systems by simplifying the matching conditions and providing flexibility in inertia forces/torques compensation. In powered exoskeletons, inertia compensation does not require modifying the mass/inertia matrix with high dimensions, and the corresponding controllers are efficient for experimental implementation. The powered exoskeletons' target control strategy uses normative torque patterns to provide assistance or augment performance. We formulate an optimization problem to design the controller and produce joint torques that fit normative biological joint torques and offload musculature for walking on multiple tasks, including stairs. We perform experiments with multiple able-bodied human subjects wearing a knee-ankle exoskeleton to demonstrate reduced activation in certain lower-limb muscles on multiple tasks.

]]>
Livestream / Virtual Mon, 21 Jun 2021 08:03:12 -0400 2021-06-22T08:00:00-04:00 2021-06-22T10:00:00-04:00 Off Campus Location Michigan Robotics Livestream / Virtual exoskeleton walking on treadmill
Mapping and Real-time Navigation with application to Small UAS Urgent Landing (July 14, 2021 11:00am) https://events.umich.edu/event/84432 84432-21623977@events.umich.edu Event Begins: Wednesday, July 14, 2021 11:00am
Location: Off Campus Location
Organized By: Michigan Robotics

Small Unmanned Aircraft Systems (sUAS) are expected to proliferate in low-altitude airspace over the coming decade requiring flight near buildings and over people.Robust urgent landing capabilities including landing site selection are required.

However, conventional fixed-wing emergency landing sites such as open fields and empty roadways are rare in and around cities. This motivates our work to uniquely consider a city's many unoccupied flat rooftops as possible nearby landing sites. We propose novel methods to identify flat rooftop buildings, isolate their flat surfaces, and find touchdown points that maximize distance to obstacles. We model flat rooftop surfaces as polygons that capture their boundaries and possible obstructions on them.

This thesis offers five specific contributions to support urgent rooftop landing. First, the Polylidar algorithm is developed which enables efficient non-convex polygon extraction with interior holes from 2D point sets. A key insight of this work is a novel boundary following method that contrasts computationally expensive geometric unions of triangles. Results from real-world and synthetic benchmarks show comparable accuracy and more than four times speedup compared to other state-of-the-art methods.

Second, we extend polygon extraction from 2D to 3D data where polygons represent flat surfaces and interior holes representing obstacles. Our Polylidar3D algorithm transforms point clouds into a triangular mesh for which planar segmentation and polygon extraction occur in parallel. Dominant plane normals are identified and used to parallelize and regularize planar segments in the mesh. Immediately after a plane has been segmented a polygon extraction task is dynamically spawned. The result is a versatile and extremely fast algorithm for non-convex polygon extraction of 3D data.

Third, we propose a framework for classifying roof shape (e.g., flat) within a city. We process satellite images, airborne LiDAR point clouds, and map building outlines to generate both a LiDAR image and a cropped satellite image of each building. Convolutional neural networks are independently trained for each modality to extract high level features and sent to a random forest classifier for final roof shape prediction. This research contributes the largest multi-city annotated dataset with over 4,500 rooftops used to train and test models. Our results show flat rooftops are identified with 90% precision and recall.

Fourth, we integrate Polylidar3D and our roof shape prediction model to reliably extract flat rooftop surfaces from archived data sources. We model risk as an innovative combination of landing site and path risk metrics and conduct a multi-objective Pareto front analysis for sUAS urgent landing in cities. We create a multi-goal planner that guarantees a risk-optimal solution is found rapidly by avoiding exploration of high-risk options. The proposed emergency planning framework enables a sUAS to select an emergency landing site and corresponding flight plan with minimum total risk.

Fifth, we verify a chosen rooftop landing site on real-time vertical approach with on-board LiDAR and camera sensors. Our method contributes an innovative fusion of semantic segmentation using neural networks with computational geometry as a hybrid algorithm robust to individual sensor and method failure. We construct a high-fidelity simulated city in the Unreal game engine with a statistically-accurate representation of rooftop obstacles. We show our method leads to greater than 4% improvement in intersection-over-union (IOU) accuracy in landing site identification compared to using LiDAR only. Finally, we evaluate results in real-world indoor flight experiments.

]]>
Lecture / Discussion Mon, 12 Jul 2021 08:03:55 -0400 2021-07-14T11:00:00-04:00 2021-07-14T13:00:00-04:00 Off Campus Location Michigan Robotics Lecture / Discussion drone determining safe touchdown site
Improving Collaboration Between Drivers and Automated Vehicles with Trust Processing Methods (July 23, 2021 9:30am) https://events.umich.edu/event/84449 84449-21624011@events.umich.edu Event Begins: Friday, July 23, 2021 9:30am
Location: Off Campus Location
Organized By: Michigan Robotics

Trust has gained attention in the Human-Robot Interaction (HRI) field, as it is considered an antecedent of people's reliance on machines.
People rely on and use machines they trust and refrain from using machines they do not trust. The advances in robotic perception technologies open paths for the development of machines that can be aware of people's trust by observing humans' behaviors and identifying whether they are being trusted or not by those people. This dissertation explores the role and the intricacies of trust in the interactions of humans and robots, particularly Automated Vehicles (AVs).
Novel methods and models are proposed for perceiving and processing drivers' trust in AVs and for determining humans' natural trust or robots' artificial trust. Two high-level problems are addressed: the problem of avoiding or reducing miscalibrations of drivers' trust in AVs and the problem of how to use trust to dynamically allocate tasks between a human and a robot that collaborate.

A complete solution is proposed for the problem of avoiding or reducing trust miscalibrations, which combines methods for estimating and influencing drivers' trust through interactions with the AV. Three main contributions stem from that solution: the characterization of risk factors that affect drivers’ trust in AVs; the development of a new method for real-time trust estimation; and the development of a new method for trust calibration.

Although the development of a complete trust-based solution for the problem of dynamically allocating tasks between a human and a robot remains an open problem, this dissertation takes a step forward in that direction. The fourth contribution here proposed is the development of a unified bi-directional model for predicting natural and artificial trust. This trust model allows for the numerical computation of human's trust and robot's trust, which is represented by the probability of a given agent to successfully execute a given task. As a probability of success, trust can readily be used for the computation of expected rewards and costs for tasks to be executed by each possible agent and can guide decision-making algorithms based on the optimization of those rewards and costs.

]]>
Lecture / Discussion Tue, 13 Jul 2021 08:09:21 -0400 2021-07-23T09:30:00-04:00 2021-07-23T11:30:00-04:00 Off Campus Location Michigan Robotics Lecture / Discussion Virtual autonomous vehicles make a left turn
Belief Representations for Planning with Contact Feedback (July 23, 2021 12:00pm) https://events.umich.edu/event/84615 84615-21624230@events.umich.edu Event Begins: Friday, July 23, 2021 12:00pm
Location: Off Campus Location
Organized By: Michigan Robotics

Chair:
Dmitry Berenson

Abstract:
While reaching for your morning coffee you may accidentally bump into the table, yet you reroute your motion with ease and grab your cup. An effective autonomous robot will need to have a similarly seamless recovery from unexpected contact. As simple as this may seem, even sensing this contact is a challenge for many robots, and when detected contact is often treated as an error that an operator is expected to resolve. Robots operating in our daily environments will need to reason about the information they have gained from contact and replan autonomously.

This thesis examines planning under uncertainty with contact sensitive robot arms. Robots do not have skin and cannot precisely sense the location of contact. This leads to the proposed Collision Hypothesis Set model for representing a belief over the possible occupancy of the world sensed through contact. To capture the specifics of planning in an unknown world with this measurement model, this thesis develops a POMDP approach called the Blindfolded Traveler's Problem. A good prior over the possible obstacles the robot might encounter is key to effective planning. This thesis develops a neural network approach for sampling potential obstacles that are consistent with both what a robot sees from its camera and what it feels through contact.

]]>
Lecture / Discussion Fri, 16 Jul 2021 09:11:47 -0400 2021-07-23T12:00:00-04:00 2021-07-23T13:30:00-04:00 Off Campus Location Michigan Robotics Lecture / Discussion Robot reaching into a fridge
Contextualized Monitoring in the Marine Environment (July 27, 2021 5:00pm) https://events.umich.edu/event/84744 84744-21624859@events.umich.edu Event Begins: Tuesday, July 27, 2021 5:00pm
Location: Off Campus Location
Organized By: Michigan Robotics

Co-Chair:
Kira Barton & Alex Shorter

Abstract:
Marine mammal monitoring has seen improvements in the last few decades with advances made to both the monitoring hardware and post-processing computation methods. However, these improvements have primarily been focused on and implemented in wild animal tracking, with less attention paid to the managed environment. This is a particularly important deficiency, as the cooperative nature of institutionally managed animals allows for research on swimming kinematics and energetics behavior with an intricacy that is difficult to achieve in the wild. This dissertation uses robotics-inspired physical kinematics and localization techniques to address two primary limitations in marine mammal monitoring: 1) the lack of persistent, absolute estimates of animal swimming energetics and kinematics, and 2) the lack of a robust, precise localization method for managed cetaceans. To resolve these, the hardware and animal tracking methods developed to enable the rest of the dissertation are first defined. Next, a physics-based approach to directly monitor cetacean swimming energetics is both presented and implemented to study animal propulsion patterns under varying effort conditions. Finally, a high-fidelity 3D monitoring framework is introduced for tracking institutionally managed cetaceans, and is applied alongside the energetics estimation method to provide a first look at the potential of spatially-contextualized animal monitoring.

]]>
Presentation Mon, 26 Jul 2021 10:13:20 -0400 2021-07-27T17:00:00-04:00 2021-07-27T19:00:00-04:00 Off Campus Location Michigan Robotics Presentation A dolphin with monitoring equipment.
Ford Motor Company Robotics Building Dedication Ceremony (September 10, 2021 3:00pm) https://events.umich.edu/event/86396 86396-21634173@events.umich.edu Event Begins: Friday, September 10, 2021 3:00pm
Location: Off Campus Location
Organized By: Michigan Robotics

Please join us as we celebrate the Ford Motor Company Robotics Building Dedication at the University of Michigan College of Engineering.

Friday, 10 September 2021
Program to begin at 3:00 PM

Michigan Robotics is embarking on several efforts to build an inclusive robotics program from the ground up, which aims to train those in the next field of engineering with concepts and values that will help enable all of humanity.

]]>
Ceremony / Service Tue, 07 Sep 2021 08:42:44 -0400 2021-09-10T15:00:00-04:00 2021-09-10T15:45:00-04:00 Off Campus Location Michigan Robotics Ceremony / Service Robotics building at night
Using Network Analysis to Understand Teamwork in Engineering: Novel Approaches, Limitations, and Future Directions (September 22, 2021 10:30am) https://events.umich.edu/event/87143 87143-21639089@events.umich.edu Event Begins: Wednesday, September 22, 2021 10:30am
Location: Off Campus Location
Organized By: Engineering Education Research

Curricular and co-curricular design experiences are an increasingly popular mechanism for delivering opportunities for students to connect technical engineering knowledge to professional skills, such as teamwork, fabrication, communication, and design ability. As such, ensuring equal participation in design activities is a critical aspect of team-based pedagogies. In this research, using data from 12 student design teams in a first-year cornerstone design course, I describe a novel approach to examining the nature of working relationships in student teams. I use the multilayer exponential random graph model (ERGM) to examine perceptions of contributions (i.e., whether one is contributing ideas to their team) and enactments (i.e., whether one’s ideas are being utilized).

The purpose of this talk is both to introduce the multilayer ERGM as novel approach to understanding teamwork in engineering, as well as to draw on this method to describe processes of inequity in undergraduate engineering education experiences. I conclude with implications for future
research and practice.

BIOGRAPHICAL SKETCH: Trevion Henderson is an Assistant Professor of Mechanical Engineering at Tufts University. He earned his Ph.D. in Higher Education from the University of Michigan, as well as his M.A. in Higher Education and Student Affairs and his B.S. in Computer Science and Engineering from The Ohio State University. Dr. Henderson holds secondary appointments in the STEM Education program in the Department of Education and the Institute for Research on Learning and Instruction (IRLI). As a first-year professor at Tufts, Dr. Henderson teaches courses on engineering design education and engineering education research methods.

]]>
Lecture / Discussion Fri, 17 Sep 2021 10:53:52 -0400 2021-09-22T10:30:00-04:00 2021-09-22T11:45:00-04:00 Off Campus Location Engineering Education Research Lecture / Discussion EER Logo
SoniCelegans - and Beyond! (September 26, 2021 2:00pm) https://events.umich.edu/event/86934 86934-21637598@events.umich.edu Event Begins: Sunday, September 26, 2021 2:00pm
Location: Earl V. Moore Building
Organized By: ArtsEngine

This is a milestone in a journey that brings together music and neurobiology. Through a collaboration among musicians, neurobiologists, sound engineers, composers, scientists, humans, and nematodes, we explore questions inspired by neurobiology experiments with the invertebrate model organism Caenorhabditis elegans (C. elegans). How does memory sound? What constitutes the musical background of a memory? How is a maze represented musically? What is the relationship between stochasticity in spatial navigation and improvisation in musical navigation? How can a human performer converse with a tiny invertebrate animal lost in space? How can we learn to create music guided by an experimental animal? How can an invertebrate creature learn to navigate guided by music? And, ultimately, how do the founding definitions of our certainties shape up when we shift our frame of thought into a new paradigm?

Sunday, September 26 | 2-3pm & 3-4pm (each show is a 30-minute installation followed by a 30-minute performance)
Admission is free but RSVP is required at https://forms.gle/AXorZMAZartHizHF8

]]>
Performance Tue, 14 Sep 2021 13:09:39 -0400 2021-09-26T14:00:00-04:00 2021-09-26T15:00:00-04:00 Earl V. Moore Building ArtsEngine Performance SoniCelegans
SoniCelegans - and Beyond! (September 26, 2021 3:00pm) https://events.umich.edu/event/86934 86934-21637599@events.umich.edu Event Begins: Sunday, September 26, 2021 3:00pm
Location: Earl V. Moore Building
Organized By: ArtsEngine

This is a milestone in a journey that brings together music and neurobiology. Through a collaboration among musicians, neurobiologists, sound engineers, composers, scientists, humans, and nematodes, we explore questions inspired by neurobiology experiments with the invertebrate model organism Caenorhabditis elegans (C. elegans). How does memory sound? What constitutes the musical background of a memory? How is a maze represented musically? What is the relationship between stochasticity in spatial navigation and improvisation in musical navigation? How can a human performer converse with a tiny invertebrate animal lost in space? How can we learn to create music guided by an experimental animal? How can an invertebrate creature learn to navigate guided by music? And, ultimately, how do the founding definitions of our certainties shape up when we shift our frame of thought into a new paradigm?

Sunday, September 26 | 2-3pm & 3-4pm (each show is a 30-minute installation followed by a 30-minute performance)
Admission is free but RSVP is required at https://forms.gle/AXorZMAZartHizHF8

]]>
Performance Tue, 14 Sep 2021 13:09:39 -0400 2021-09-26T15:00:00-04:00 2021-09-26T16:00:00-04:00 Earl V. Moore Building ArtsEngine Performance SoniCelegans