View this email in your browser
Visit us on the web

Updates from IRIM | September 2, 2021

Fall 2021 Industry & Student Mixer
September 15, 2021 | 10am - 12pm

Georgia Institute of Technology
Klaus Advanced Computing Building: 1116
266 Ferst Drive | Atlanta, GA 30332-3000

We hope you plan to join us in person and on campus at this year’s annual Industry & Student Mixer on Tuesday, September 15, 2021, from 10AM - 12PM EDT. In addition to our current and incoming Ph.D. students, industry participants will have the opportunity to meet the inaugural M.S. in Robotics cohort.

Industry Guests: Click Here to Register

GT Students: Click Here to Register

Robotics News

Georgia Tech Joins the U.S. National Science Foundation to Advance AI Research and Education

NSF AI Institute for Collaborative Assistance and Responsive Interaction for Networked Groups (AI-CARING) will seek to create a vibrant discipline focused on personalized, collaborative AI systems that will improve quality of care for the aging. The systems will learn individual models of human behavior and how they change over time and use that knowledge to better collaborate and communicate in caregiving environments. Led by Sonia Chernova, associate professor of interactive computing at Georgia Tech, the AI systems will help a growing population of older adults sustain independence, improve quality of life, and increase effectiveness of care coordination across the care network.

“The AI-CARING Institute builds on our existing strengths in AI and in technology for aging. It will create not only novel solutions, but a new generation of researchers focused on the interaction between the two,” said Charles Isbell, dean and John P. Imlay Jr. Chair in the College of Computing. “Our aim is to build cutting-edge technologies that improve the lives of everyone, and I can’t think of a better example than AI-CARING.”

Read the Full Announcement Here

Simple Linking of Units Gives Legged Robots New Way to Navigate Difficult Terrain

At the opening ceremony of this summer’s Tokyo Olympics, a fleet of 1,824 drones flew above the stadium, illuminating the night with an environmentally friendly light show. Such displays of reconfigurable, floating lights have one important common factor: a crash- and stumble-free navigable area in which to perform. The success of robotic swarms in aerial, aquatic, and land-based environments can be attributed to ease of navigation in a homogenous or highly controlled space.

But, what about more complex terrain? The capability that would allow land-based search-and-rescue robot swarms to navigate buildings and other disaster areas does not yet exist. Researchers at the Georgia Institute of Technology are working to develop simple, low-cost, legged robots capable of linking and unlinking to accomplish tasks, such as gap traversal, stair climbing, and object transport over uneven terrains.

Working with Daniel Goldman, Dunn Family professor in the School of Physics at Georgia Tech, Yasemin Ozkan-Aydin, a former postdoc in Goldman’s lab and now an assistant professor at the University of Notre Dame, developed “quadruped” robots using easily acquired off-the-shelf technology.

Read the Full Article Here

Robotics REU Student Spotlight

A Summer for Path Planning Lessons: for Robots & REU Student Alike

Established as part of the National Science Foundation (NSF) Research Experiences for Undergraduates (REU) Summer Undergraduate Research in Engineering (SURE) grant, the SURE Robotics Program adds a robotics component to Georgia Tech’s SURE Program, established in 1992.

Over the months of the 2021 Fall Semester, IRIM will be highlighting each of the  undergraduate participants, their research topics and experience in the labs, as well as what they gained from the program and their time at Georgia Tech, and in Atlanta. Our first interviewee from the program is Mark Jimenez, an Incoming Senior in Computer Science at the University of Hawai`I at Hilo.

IRIM 2021 SURE REU Student Mark Jimenez in the DART Lab IRIM 2021 SURE REU Student Mark Jimenez in the DART Lab Name: Mark Jimenez
PI: Prof. Anirban Mazumdar 
Mentor: Hayden Nichols

1. What sparked your interest in robotics and what problems are you hoping to help solve as a roboticist?

My interest in robotics stems from my interest in Artificial Intelligence (AI). While following my interest in AI and computing, I found that many difficult and important problems in robotics are in the field of autonomy, and AI is looked to as a key in much of the research in this field. I personally believe that robotics is one of the most obvious and direct ways in which AI can benefit humans and our society. From caretakers and autonomous vehicles to household and industrial robots, AI and robotics must be paired in order to take on complex problems. More importantly than simply teaching robots to do work that humans can, I see robotics as expanding access to critical services such as healthcare, transportation, food distribution, and emergency services. Moreover, I believe that robotics can and should be used to make the lives of working people simpler and safer, by increasing the use and efficacy of reliable autonomous systems in factories, small businesses, and homes. Personally, I hope to help solve the problem of robot education via "self-direction" advancing robotic learning abilities so that our robots are malleable enough to know how to teach themselves new skills, rather than needing to be reprogrammed every time a new task presents itself.

2. What research are you conducting at GT and what applications do you feel this research may have?

At GT, I am working on mathematically based robotic path planning (A to B) and classification (of the robot's surrounding environment). There are many obvious applications of path planning, and planning algorithms have been implemented extensively in real‐world robotic systems, from welding robots to self‐driving cars. I find the classification aspect of this project to be particularly intriguing, as many robots currently in use are completely unaware of their environmental surroundings, outside of their programmed task. Having robots that can anticipate the movement of other agents in their environment is an important step in expanding the use of robots in household and "human" environments such as hospitals and care facilities, or hazardous search and rescue tasks. Our project examines both path planners and observers.  The planners are creating trajectories and the observers are trying to understand the planner behavior. This relationship can be symbiotic robots predicting each other's paths and future positions could be a critical aspect of many cooperative robotic tasks. For example, imagine a cooking and cleaning robotic pair, explicit cooperative planning may not be feasible, and so these robots would need to accomplish their tasks without interfering with one another’s work.

3. What has been your favorite academic lab activity/ tool training/ etc. thus far and why?

Being able to build multiple neural networks, each individually capable of time series classification in order to solve an important problem, has been my favorite activity. I have worked with neural networks in the past, but building a network from scratch and at a low level, using TensorFlow and not some higher level API has been a tremendously valuable learning experience. Writing at this lower level has allowed me to understand more fully what is happening, programmatically, during the training of a neural network. Moreover, having access to the underlying operations taking place during training and prediction allows me to create highly customizable networks and layers in future research.

4. Do you feel this REU experience has helped prepare you for working in a collaborative laboratory environment and furthered your education goals?

This REU has undoubtedly prepared me for working in a collaborative research environment; I have been able to cooperate with both master’s and Ph.D. students, as well as a professor/lab director, so I have been working alongside every major academic role in a university lab. More importantly, I have been able to learn valuable insights and skills from each of these academics, and I have a much better understanding of how high‐level research laboratories function and produce work. This has unquestionably furthered my academic goals, as I have been able to work full‐time in a research setting on an incredibly interesting project with many avenues for exploration, which is exactly what I want to do.

5. What are your plans post-undergraduate?

After graduating from my current program in Computer Science, I intend to obtain a Ph.D. in Computer Science, with a focus in Artificial Intelligence. To me, robotics is one of the most exciting aspects of Artificial Intelligence, and robot learning is something of paramount interest to me. Moreover, the history of AI research in robotics has shown how difficult and data‐intensive robot learning is, meaning that there is considerable work and progress to be made in this field. I believe AI has tremendous implications for robotics, and robotics is the field in which AI can have an immediate and profound impact on our society and in our day to day lives, which is exactly what excites and drives me to pursue a Ph.D. in AI.

6. What is/was your favorite thing about/impression of GA Tech and ATL?

GT has a beautifully green campus, exceptional facilities, innovative ideas, and hard‐working, passionate researchers. Specifically in my lab, the students here have been extremely welcoming and helpful, with genuine passion and excitement for their research. It has been a joy to come into work each day and work alongside such enthusiastic and talented individuals. Atlanta is likewise a beautiful setting for research, with comfortable weather, rich natural beauty, and an original, diverse culture.
Upcoming Events

Shedding Light on 3D Cameras
September 8, 2021 | 12:15 - 1:15 PM EST

Klaus Advanced Computing Building 1116 | 266 Ferst Dr NW
Featuring Mohit Gupta | Assistant Professor; Computer Sciences, University of Wisconsin-Madison

Abstract: The advent (and commoditization) of low-cost 3D cameras is revolutionizing many application domains, including robotics, autonomous navigation, human computer interfaces, and recently even consumer devices such as cell-phones. Most modern 3D cameras (e.g., LiDAR) are active; they consist of a light source that emits coded light into the scene, i.e., its intensity is modulated over space, and/or time. The performance of these cameras is determined by their illumination coding functions.

I will talk about our work on developing a coding theory of active 3D cameras. This theory, for the first time, abstracts several seemingly different 3D camera designs into a common, geometrically intuitive space. Based on this theory, we design novel 3D cameras that achieve up to an order of magnitude higher performance as compared to the current state-of-the art. I will also briefly talk about our work toward developing `All-Weather’ 3D cameras that can operate in extreme real-world conditions, including outdoors (e.g., a robot navigating outdoors in bright sunlight and poor weather), under multi-camera interference (e.g., multiple robots navigating in a shared space such as a warehouse), and handle optically challenging objects such as shiny metal (e.g., for an industrial robot sorting machine parts).

Bio: Mohit Gupta is an Assistant Professor of Computer Sciences at the University of Wisconsin-Madison. He received B. Tech in Computer Science from IIT-Delhi, Ph.D. from the Robotics Institute, Carnegie Mellon University, and was a postdoctoral research scientist at Columbia University. He directs the WISION Lab with research interests broadly in computer vision and computational imaging. He has received best paper honorable mention awards at computer vision and photography conferences in 2014 and 2019, and a Sony Faculty Innovation Award in 2019. His research is supported by NSF, ONR, DARPA, Sony, Intel and Wisconsin Alumni Research Foundation.

Updated COVID News & Guidance

🏆 This week, Georgia Tech is introducing new rewards to encourage individuals to get vaccinated and test weekly for Covid-19. Items include gift cards, t-shirts, signed footballs and helmets, discounts at local restaurants, and more.

Additionally, to encourage unvaccinated USG employees to schedule their Covid-19 vaccination appointment, and to thank those who are already vaccinated, the USG will provide employees with eight hours of administrative leave.

🤒 An Update on Covid-19 Symptoms and Cases 
Many cases of Covid-19 may present as illness with mild symptoms — a runny nose, a sore throat, a headache. Dr. Benjamin Holton, senior director at Stamps Health Services, shares what he's seen and what action you should take if you have symptoms.

🐝 Tech Moving Forward
Everyone in the Georgia Tech community is encouraged to follow the CDC's recommendations and wear a mask in campus buildings. Learn more about campus updates.

Institute for Robotics and Intelligent Machines (IRIM)
Klaus Advanced Computing Building
266 Ferst Drive
Atlanta, GA 30332-3000

Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.

Copyright © 2021 Institute for Robotics and Intelligent Machines (IRIM) at Georgia Tech
All rights reserved.