Category Archives: News

Milking cows for data – not just dairy products

Optimizing cows.

 

In the mid-1970s, the average American dairy farm had about 25 cows. Today, many operations have more than 3,000 – a number that was almost unheard of 25 years ago.

Managing large herds efficiently would be difficult, perhaps even impossible, without the latest advances in computing and automation. Most dairies now have milking parlors and associated free-stall housing, which double or triple production per man-hour. Milking units automatically detach to reduce udder health problems and improve milk quality, while cow ID transponders let farmers automatically record production data.

The most recent major technological advance influencing the U.S. dairy industry is the development of automatic milking systems – or “robotic” milkers.

At the University of Connecticut’s Kellogg Dairy Center, we’re using robotic milkers as well as other sensors to monitor 100 cows and their physical environment. Through this work, launched this spring, we hope to monitor individual cow’s behavior and health in real time to improve production efficiency and animal well-being.

Big data and cows

Robotic milkers can harvest milk without human involvement. In fact, the cows decide when to be milked, entering the machine without direct human supervision. The robotic system automatically identifies the cow and applies a sanitizing teat spray before a robotic arm attaches the teat cup for milking.

That’s very different from parlor milking, where managers decide when to milk cows, usually three times a day. Each robotic milking unit serves 50 to 55 cows.

Given the high price of early versions of the robotic milkers and the large size of U.S. herds, American dairies had minimal interest in robotic milkers before 2010. However, the number of automatic milking systems in the country increased to over 2,500 units in 2013, mainly due to improvements in design in the newer models. Worldwide, there are currently over 35,000 automatic milking systems in operation.

A row of cows being milked

Not only have these newer machines improved in harvesting milk efficiently, they have the added ability to collect a greater amount of information about production, milk composition and cow behavior. That allows producers to make more informed management decisions.

With robotic milking systems, the cows run the show. They decide when to eat, ruminate, rest or be milked. They also need to spend less than an hour per day actually being milked; before robotic milkers, milking often took up three to five hours per day.

We wanted to know: What are they doing with the rest of their day? How does that behavior affect production or serve to indicate health status? By themselves, the milking units can’t gather that kind of information, which would be very useful in finding out early on whether a particular cow is developing a health problem.

Our “cow-CPS” – a cyber-physical system that includes the cows, robotic milkers, video cameras and other sensors – will track data on our cows at all times. That will tell us, among other things, where the cows go when not being milked; when they decide to eat, rest or do other activities; and the composition of their milk. Sensors placed inside the body will even tell us the pH inside one of their stomachs, which could be a key indicator of any digestive problems.

Optimizing dairies

We hope that all of this data will allow us to make timely decisions at the level of the individual cow, something that’s not easy to do in large herds. This “precision dairying” could help us understand how an individual cow’s activities – eating, standing, resting, milking – affects her milk production, milk quality and health.

We plan to analyze the data with the help of machine learning, a type of artificial intelligence that can find patterns in large amounts of information. The computer will compare the data against a model of how the dairy should operate under ideal conditions. Our model captures critical performance characteristics – milk quality and productivity – as well as relevant constraints, such as individual health and reproductive status.

As the dairy operates, the real-time data will allow us to assess how far away our real farm is from the ideal one. We can then combine this information with a mathematical optimization algorithm to determine how exactly we should modify or adjust the process. For example, the algorithm may suggest adjusting the type of teat drip, the nutritional content of the feed or the amount of time each cow spends feeding.

We hope that our work will allow dairy farmers across the U.S. to better manage individual cows in a group setting – not only to improve milk production, but to bolster cow health.

Javidi Wins Prestigious Award from The Optical Society

The UConn School of Engineering is pleased to announce that Dr. Bahram Javidi, Board of Trustees Distinguished Professor in Electrical and Computer Engineering, has been awarded the prestigious Joseph Fraunhofer Award / Robert M. Burley Prize from The Optical Society.

The award, given to one person in the country every year, was bestowed upon Javidi for his “seminal contributions to passive and active multi-dimensional imaging from nano- to micro- and macro-scales,” according to the award citation.

The award is one of a long line of accomplishments during his career, which include: Being named one of the top 160 engineers between the ages of 30-45 by the National Academy of Engineering (NAE); the Quantum Electronics and Optics Prize for Applied Aspects by the European Physical Society; the Dennis Gabor Award in Diffractive Wave Technologies from The International Society for Optics and Photonics (SPIE); the John Simon Guggenheim Foundation Fellowship; the Alexander von Humboldt Prize for senior US Scientists in all disciplines; the SPIE Technology Achievement Award; the National Science Foundation Presidential Young Investigator Award; and the George Washington University Distinguished Alumni Scholar Award.

At UConn, he has received the American Association for University Professors (AAUP) Research Excellence Award; the University of Connecticut Board Of Trustees Distinguished Professor Award; the UConn Alumni Association Excellence in Research Award; and the Chancellor’s Research Excellence Award, among others.

He is a Fellow of the Institute of Electrical and Electronics Engineers (IEEE), Fellow of the American Institute for Medical and Biological Engineering (AIMBE), Fellow of the Optical Society of America (OSA), Fellow of the European Optical Society (EOS), Fellow of The International Society for Optics and Photonics (SPIE), Fellow of the Institute of Physics (IoP), and Fellow of The Society for Imaging Science and Technology (IS&T). Javidi has over 900 publications and 19 patents, some of which have been licensed by industry.

Javidi is also the director of the MOSIS Lab (Multidimensional Optical Sensing and Imaging Systems), which is focused on advancing the science and technology of imaging, by centering on the fields of optics, photonics, and computational algorithms and systems, from nano to macro scales. MOSIS works with, and finds solutions for, partners in the defense, manufacturing, healthcare, and cybersecurity industries.

Click here to learn more about the Joseph Fraunhofer Award / Robert M. Burley Prize from The Optical Society.

Remembering Dr. John Enderle, The Educator and The Nurturer


Dr. John Enderle and his wife, Laurie Enderle. (photo courtesy of Laurie Enderle)

Originally by: Heidi Douglas, Director of Alumni Relations, UConn School of Engineering 

Professor Emeritus John D. Enderle, 65, of Ashford, CT passed away on April 2, 2018 after a long and courageous battle with pancreatic cancer.  A loving husband, father, brother, friend, colleague and mentor, Dr. Enderle enjoyed a long and illustrious career as a professor and inspirational leader, admired for his unfailing dedication and support for students, a legacy honored by his family establishing the John Enderle Fund memorial scholarship.

UConn Electrical and Computer Engineering department head from 1995-1997, John was founding director of the undergraduate Biomedical Engineering program in 1997. His passion for research and advising his students are storied with commendations describing him as “the greatest professor,” having a “major influence in my life over the past 20 years,” and “always had patience to help me pursue my goals.”

John earned his B.S., M.E., and Ph.D. degrees in Biomedical Engineering, and an M.E. degree in Electrical Engineering from Rensselaer Polytechnic Institute. He worked at the National Science Foundation (NSF) and was as a professor at North Dakota State University (NDSU) prior to joining UConn.

In addition to his teaching and research, John also served in many capacities for several professional societies, was a member of the Connecticut Academy of Science and Engineering, a former Accreditation Board for Engineering and Technology (ABET) Program Evaluator for Bioengineering Programs and member of the Engineering Accreditation Commission. He was Editor of the NSF Book Series on NSF Engineering Senior Design Projects to Aid Persons with Disabilities. At the time of his death, John was working on a fourth edition of his seminal undergraduate textbook for biomedical engineering, Introduction to Biomedical Engineering.

A particularly metaphorical tribute to John celebrates his passion for gardening. A painting by former student Dr. G. Alexander Korentis depicts an espalier apple tree with John’s name followed by all his Ph.D. students’ names in an upward succession of branches. The frame bears a plaque with a quote from Warren Buffett, “Someone is sitting in the shade today because someone planted a tree a long time ago.”

John Enderle planted a tree for all of his students.

 Donations may be made in memory of John Enderle to the “John Enderle Fund” in the UConn School of Engineering. Please make checks payable to: The UConn Foundation, Inc. and forward to the following address: 2390 Alumni Drive Unit 3206, Storrs, Connecticut 06269.

Senior Design: Building The “Heart and Soul” of an Electric Car


Ortega-Hernandez, Tshipamba, and Biron look over their EMRAX motor in the Castleman Building Machine Shop (Christopher Larosa/UConn Photo)

Twenty years ago, if you stood on a sidewalk and watched cars go by, chances are high that you would see little-to-no electric cars driving down the street. In 2017, electric car sales were higher than ever, with nearly 200,000 all-electric cars sold in the U.S. With the popularity of models from Tesla, BMW, and Chevy, consumers are starting to warm to the idea of charging their car, rather than filling it with gasoline. Because of that popularity, four Senior Design teams, including an electrical and computer engineering team featuring seniors Daryl Biron; Ernesto Ortega-Hernandez; and Alain Tshipamba, are working to complete an all-electric car for a national competition in June.

The portion of the car that Biron, Ortega-Hernandez, and Tshipamba are working on is the “heart and soul” of the vehicle—the powertrain. The sponsor of the project, the UConn Electric Motorsports, was originally formed in the spring of 2017, with the intention of getting like-minded students together to build a car that could compete in Formula North, a collegiate competition taking place during in the summer of 2018. The advisor of the team is Professor Ali Bazzi. 

Biron said he originally got involved after having interest in the club in the spring 2017 semester:

“I got involved with the club personally, in the spring semester last year, and at that time there were only two electrical engineers involved, and with an electric car, you need a lot more than just two,” Biron said. “So, I was pretty much thrown onto the powertrain team, which is essentially everything that the motor controls, and I didn’t really know much, so I had to do a lot of research on my own, which became easier when Ernesto and Alain came onboard.”


The EMRAX motor provides 80 kilowatts of power, equivalent to 107 horsepower (Christopher Larosa/UConn Photo)

The car itself will have a chassis made of inch-thick aluminum honeycomb sheets, which will make it one of the lightest and most torsionally rigid chassis seen in completion, according to the UCEM website. The car will also use pieces and materials that will make the car extremely flexible and ergonomic, with components like adjustable pedals and a removable seat.

The permanent magnet motor, which the group only received recently, after a wait of a few months, was designed by EMRAX, and provides 80 kilowatts of power, equivalent to 107 horsepower. Ortega-Hernandez said they had to jump through a lot of hoops before the motor arrived at its destination:

“Initially we had some funding problems, which were later solved, but when we went to order the motor, we ran into issues, because it was coming from Europe, EMRAX required a wire transfer for payment, and they weren’t an approved UConn vendor. So, luckily, EnviroPower, the company where Daryl interns, offered to become a vendor, and then ordered the motor through their channels—but when all was said and done, the entire purchasing process took three months.”

The rest of the powertrain consists of an emDrive300H Controller from Emsiso, and several other components, which will connect to a battery apparatus being constructed by another ECE team.

Tshipamba also said that getting all the calculations fined-tuned was one of their biggest early challenges:

“Luckily, when we went to our advisor, he really helped us find out what we were doing wrong. Originally, we had issues with the simulation models, which were due to using parts from different libraries that didn’t communicate well, and were also not adjusted to our parameters regardless of tuning, so we realized that we had to create our own parts based on the mathematical model.”

But Biron said that a lot of these impediments eventually turned into worthwhile accomplishments:

“Actually, getting the motor and fixing our mathematical models was a huge breakthrough for us,” Biron said. “At one point, we were talking about ordering the motor the beginning of November, but obviously things got in the way.”

Now that they have the motor, all their components, and all of the modeling squared away, the group is looking forward to getting to work, and putting their focus towards putting all the pieces together for April and beyond.

 


Biron holds the EMRAX motor, which sits next to the Emsiso emDrive300H Controller. (Amanda Wright/UConn Photo)

With good intentions in mind, the Electrical and Computer Engineering Senior Design team of Daryl Biron, Ernesto Ortega-Hernandez, and Alain Tshipamba set out to build not only the powertrain of an electric car, but also realize their dream of seeing it race in the summer.

The portion of the car that Biron, Ortega-Hernandez, and Tshipamba are working on is the “heart and soul” of the vehicle—the powertrain. The sponsor of the project, the UConn Electric Motorsports Club, was originally formed in the spring of 2017, with the intention of getting like-minded students together to build a car that could compete in Formula North, a collegiate competition taking place during in the summer of 2018. The advisor of the team is Professor Ali Bazzi. 

Unfortunately, due to multiple factors, Biron said that the car will not be ready for Formula North this summer:

“There’s a lot that has gone into this car, and funding has been an issue, so even this upcoming summer, when Senior Design finishes, the club probably won’t be closing in on finishing the car, but will be thinking about the overall design, and how we can make it better,” Biron said. “So, odds are it won’t be fully built until January 2019.”

When the car is officially completed, it will have a chassis made of inch-thick aluminum honeycomb sheets, which will make it one of the lightest and most torsionally rigid chassis seen in competition, according to the UCEM website. The car will also use pieces and materials that will make the car extremely flexible and ergonomic, with components like adjustable pedals and a removable seat.

The permanent magnet motor, which the group only received in February, after a wait of a few months, was designed by EMRAX, and provides 80 kilowatts of power, equivalent to 107 horsepower.


Biron holds the EMRAX motor, which sits next to the Emsiso emDrive300H Controller. (Amanda Wright/UConn Photo)

The rest of the powertrain consists of an emDrive300H Controller from Emsiso, and several other components, which will connect to a battery apparatus being constructed by another ECE team. Unlike the ordering of the motor, which was long and drawn out, Biron said the ordering of the controller was extremely easy:

“The ordering of the motor controller was much smoother, and it actually came in a couple of weeks earlier than we anticipated,” Biron said. “The only problem we encountered was that on the website, it said we would be able to use any USB-to-CAN adapter to connect to our computer to program it, but after we ordered it we found out that we needed to order a special adapter from the company.”

Biron said if they used a standard adapter, then they wouldn’t have been able to see any of the real-time graphs or data logging, so they attempted to jerry-rig it, but, unfortunately, they found out that they ultimately need to buy the special adapter.  

But the powertrain, which was the team’s focus for Senior Design, will absolutely be ready to go for Senior Design Demonstration Day on April 27, according to Biron:

“We’re going to put all the parts completely together into one system before Senior Design Day, and then we’ll begin testing,” Biron said. “Then, since I don’t think we’ll be able to integrate the system into the car, we’ll have to work on a test demo for Senior Design Day, showing how we control the motor and get it to spin correctly.”

Asked about the whole year-long process, Ortega-Hernandez said that it’s definitely been a long road, but has also been a great experience:

“I think I’ll feel accomplished when I actually see that motor spin, and we confirm that we’re seeing the correct speed and torque,” Ortega-Hernandez said. “It’s been a frustrating and tough process, but we’re definitely proud of the work we’ve put in and the final result.”

The team’s final product will be presented on Senior Design Day, April 27, from 1-4 p.m., in Gampel Pavillion. To attend the event, please RSVP by clicking this link.

Designing a Smart Sensor Network for Tracking Submarines

Illustration of network concept. A UConn researcher at the National Institute for Undersea Vehicle Technology is developing a ‘smart sensor network’ that is both energy-efficient and resilient, to track targets such as enemy submarines. (Getty Images)


A team of UConn engineers is developing an energy-efficient “smart sensor network” to track targets of interest, such as the proximity of enemy submarines or ships to Navy vessels.

The U.S. Navy currently uses underwater Intelligence, Surveillance, & Reconnaissance (ISR) sensor networks that run on full power, which can be a problem for long-term operations. The more accurate the sensor, the more power they consume.

The sensor networks currently being used could consist of several multi-modal sensor nodes, called sensor buoys, where each node acts independently and contains a diverse sensor suite, a data-processing unit, a transmitter and receiver, and a GPS device. The sensor suite can be composed of different types of sensors to detect and track targets, such as underwater microphones and active sonars.

Traditionally, these sensor nodes operate on full power, running all devices simultaneously, but the batteries that power them typically burn out within a few days of operation,  just as cell phones suck up more power when running multiple operations. This causes sensing failures which, in turn, leads to holes in coverage and affects tracking performance.

This poses a challenge to the Navy, since it deploys thousands of acoustic sensor networks throughout the ocean, where battery replacement can be time-consuming or impossible.

To address the challenge, Shalabh Gupta, a UConn engineer and researcher at the National Institute for Undersea Vehicle Technology, devised the concept of a “smart sensor network” that is energy-efficient as well as resilient to failures.

 Intelligent Energy-efficient Sensor Network. (Illustration by Hayley Joyal ’18 (SFA))

In a smart sensor network, sensor nodes adapt their sensing modalities based on the information about the targets’ whereabouts. Thus, the nodes around the target, such as a ship or submarine, activate their high-power sensing devices to track the target accurately, pinpointing its location, velocity, and trajectory.

On the other hand, the nodes that are located farther away from the target cycle between low-power sensing and sleep states to minimize energy consumption while still remaining aware.

Thus, if a low-power sensor detects a target, the node switches to high-power sensing to track it. Similarly, the high-power sensing devices that are tracking the target predict the target’s trajectory and alert other sensors within range of the target’s path, so that they switch to high power. Once the target has passed outside of a sensor’s range, it reverts to low-power mode.

The smart sensor networks also provide resilience. If a few nodes in the network fail, then the nodes surrounding the hole in coverage formed by the failed nodes jointly optimize to expand their sensing ranges to cover the gap.

“These networks have to contain built-in, distributed intelligence,” says Gupta, an assistant professor of electrical and computer engineering.

His first research paper on the algorithm, coauthored by graduate student James Hare, was published online in IEEE Transactions on Cybernetics in August 2017.

With this advance, crews on ships and submarines will be able to track enemy watercraft with batteries that last about 60 to 90 percent longer, Gupta says.

Gupta’s lab has prototypes of the sensors for ground use, and has been talking with Navy personnel about using them for the underwater acoustic sensor network.  He is currently seeking funding to build underwater sensors.

How UConn Researchers are Teaching Robots to Think Like Humans

There’s a great scene in the movie “Iron Man” where Robert Downey Jr.’s character Tony Stark (aka Iron Man) is crawling across his lab, desperately trying to reach the small arc reactor he needs to keep his heart beating and stay alive.

Weakened by a run-in with arch villain Obadiah Stane, Stark can’t reach the gizmo where it sits on a tabletop. Defeated, he rolls onto his back, exhausted and pondering his inevitable doom.

But the very moment that we think our intrepid hero’s a goner, a metallic hand appears at Stark’s shoulder, holding the lifesaving device. “Good boy,” Stark says weakly as he takes the device from his robot assistant, Dum-E.

And just like that, our hero is saved.

From the dutiful shuffling of C-3PO to the terrorizing menace of The Terminator, Hollywood has made millions tantalizing audiences with far-out robot technology. Scenes like the one in “Iron Man” make for good entertainment, but they also are based, to some degree, in reality.

Dum-E’s interaction with Stark is called collaborative robotics, where robots with advanced artificial intelligence, or A.I., not only work alongside us humans but also are able to anticipate our actions and even grasp what we need.

Collaborative robotics represents the frontier of robotics and A.I. research today. And it’s happening at UConn.

Three thousand miles away from the klieg lights of Hollywood, Ashwin Dani, director of UConn’s Robotics and Controls Lab, or RCL, stands in the stark fluorescent light of his Storrs office staring at a whiteboard covered in hastily scrawled diagrams and mathematical equations.

Here, in the seemingly unintelligible mishmash of numbers and figures, are the underlying mathematical processes that are the lifeblood of collaborative robotics.

If robots are going to interact safely and appropriately with humans in homes and factories across the country, they need to learn how to adapt to the constantly changing world around them, says Dani, a member of UConn’s electrical and computer engineering faculty.

“We’re trying to move toward human intelligence. We’re still far from where we want to be, but we’re definitely making robots smarter,” he explains.

All of the subconscious observations and moves we humans take for granted when we interact with others and travel through the world have to be taught to a robotic machine.

When you think about it, simply getting a robot to pick up a cup of water (without crushing it) and move it to another location (without spilling its contents or knocking things over) is an extraordinarily complex task. It requires visual acuity, a knowledge of physics, fine motor skills, and a basic understanding of what a cup looks like and how it is used.

“We’re teaching robots concepts about very specific situations,” says Harish Ravichandar, the senior Ph.D. student in Dani’s lab and a specialist in human-robot collaboration. “Say you’re teaching a robot to move a cup. Moving it once is easy. But what if the cup is shifted, say, 12 inches to the left? If you ask the robot to pick up the cup and the robot simply repeats its initial movement, the cup is no longer there.”

Repetitive programs that work so well for assembly-line robots are old school. A collaborative robot has to be able to constantly process new information coming in through its sensors and quickly determine what it needs to do to safely and efficiently complete a task. If that robot is part of an assembly line, the line has to shut down and the robot has to be reprogrammed to account for the change, an inefficient process that costs manufacturers money. Hence the thinking robot this team is trying to create.

While the internet is filled with mesmerizing videos of robots doing backflips, jumping over obstacles, and even making paper airplanes, the UConn team’s effort at controlling robots through advanced artificial intelligence is far less flashy but potentially far more important.

Every move the UConn team wants its test robot to make starts here, says Dani, with control theory, engineering, whiteboards, and math.

“We’re writing algorithms and applying different aspects of control theory to take robot intelligence to a higher level,” says Ravichandar. “Rather than programming the robot to make one single movement, we are teaching the robot that it has an objective — reaching for and grabbing the cup. If we succeed, the robot should be able to make whatever movements are necessary to complete that task no matter where the cup is. When it can do that, now the robot has learned the task of picking something up and moving it somewhere else. That’s a very big step.”

While most of us are familiar with the robots of science fiction, actual robots have existed for centuries. Leonardo da Vinci wowed friends at a Milan pageant in 1495 when he unveiled a robotic knight that could sit, stand, lift its visor, and move its arms. It was a marvel of advanced engineering, using an elaborate pulley and cable system and a controller in its chest to manipulate and power its movements.

But it wasn’t until Connecticut’s own Joseph Engelberger introduced the first industrial robotic arm, the 2,700-pound Unimate #001, in 1961 that robots became a staple in modern manufacturing.

Unimates were first called into service in the automobile industry, and today, automobile manufacturers like BMW continue to be progressive leaders using robots on the factory floor. At a BMW plant in Spartanburg, South Carolina, for example, collaborative robots help glue down insulation and water barriers on vehicle doors while their human counterparts hold the material in place.

The advent of high-end sensors, better microprocessors, and cheaper and easily programmable industrial robots is transforming industry today, with many mid-size and smaller companies considering automation and the use of collaborative robots.

Worldwide use of industrial robots is expected to increase from about 1.8 million units at the end of 2016 to 3 million units by 2020, according to the International Federation of Robotics. China, South Korea, and Japan use the most industrial robots, followed by the United States and Germany.

Anticipating further growth in industrial robotics, the Obama administration created the national Advanced Robotics Manufacturing Institute, bringing together the resources of private industry, academia, and government to spark innovations and new technologies in the fields of robotics and artificial intelligence. UConn’s Robotics and Controls Lab is a member of that initiative, along with the United Technologies Research Center, UTC Aerospace Systems, and ABB US Corporate Research in Connecticut.

Manufacturers see real value in integrating collaborative robots into their production lines. The biggest concern, clearly, is safety.

There have been 39 incidents of robot-related injuries or deaths in the U.S. since 1984, according to the federal Occupational Safety and Health Administration. To be fair, none of those incidents involved collaborative robots and all of them were later attributed to human error or engineering issues.

The first human known to have been killed by a robot was Robert Williams in 1979. Williams died when he got tired of waiting for a part and climbed into a robot’s work zone in a storage area in a Ford Motor plant in Flat Rock, Michigan. He was struck on the head by the robot’s arm and died instantly. The most recent incident happened in January 2017, when an employee at a California plastics plant entered a robot’s workspace to tighten a loose hose and had his sternum fractured when the robot’s arm suddenly swung into action.

“When you have a human and a robot trying to do a joint task, the first thing you need to think about of course is safety,” says Dani. “In our lab, we use sensors that, along with our algorithms, not only allow the robot to figure out where the human is but also allow it to predict where the human might be a few seconds later.”

One way to do that is to teach robots the same assembly steps taught to their human counterparts. If the robot knows the order of the assembly process, it can anticipate its human partners’ next moves, thereby reducing the possibility of an incident, Dani says. Knowing the process would also allow robots to help humans assemble things more quickly if they can anticipate an upcoming step and prepare a part for assembly, thus improving factory efficiency.

“Humans are constantly observing and predicting each other’s movements. We do it subconsciously,” says Ravichandar. “The idea is to have robots do the same thing. If the robot sees its human partner performing one step in an assembly process, it will automatically move on to prepare for the next step.”

Which brings us back to the whiteboards. And the math.

Failure is always an option. But when the math finally works, Ravichandar says, the success is exhilarating.

“Once you have the math figured out, it’s the best feeling because you know what you want the robot to do is going to work,” Ravichandar says with an excited smile.

“Implementing it is a whole other challenge,” he adds quickly, his passion for his work undiminished. “Things never work the first time. You have to constantly debug the code. But when you finally see the robot move, it is great because you know you have translated this abstract mathematical model into reality and actually made a machine move. It doesn’t get any better than that.”

With an eye on developing collaborative robotics that will assist with manufacturing, Dani and his team spent part of the past year teaching their lab’s test robot to identify tools laid out on a table so it can differentiate between a screwdriver, for example, and a crescent wrench, even when the tools’ initial positions are rearranged. Ultimately, they hope to craft algorithms that will help the robot work closely with a human counterpart on basic assembly tasks.

Another member of the team, Ph.D. candidate Gang Yao, is developing programs that help a robot track objects it sees with its visual sensors. Again, things we humans take for granted, such as being able to tell the difference between a bird and a drone flying above the trees, a robot has to learn.

Building advanced artificial intelligence doesn’t happen overnight. Ravichandar has been working on his projects for more than three years. It is, as they say, a process. Yet the team has learned to appreciate even the smallest of advances, and late last year, he flew to California to present some of the lab’s work to an interested team at Google.

“C-3PO is a protocol droid with general artificial intelligence,” says Ravichandar. “What we are working on is known as narrow artificial intelligence. We are developing skills for the robot one task at a time and designing algorithms that guarantee that whatever obstacles or challenges the robot encounters, it will always try to figure out a safe way to complete its given task as efficiently as it can. With generalized intelligence, a robot brings many levels of specific intelligence together and can access those skills quickly on demand. We’re not at that point yet. But we are at a point where we can teach a robot a lot of small things.”

Inevitably, as robots gain more and more human characteristics, people tend to start worrying about how much influence robots may have on our future.

Robots certainly aren’t going away. Saudi Arabia recently granted a robot named Sophia citizenship. Tesla’s Elon Musk and Deep Mind’s Mustafa Suleyman are currently leading a group of scientists calling for a ban on autonomous weapons, out of concern for the eventual development of robots designed primarily to kill.

Although it doesn’t apply directly to their current research, Dani and Ravichandar say they are well aware of the ethical concerns surrounding robots with advanced artificial intelligence.

Ravichandar says the problem is known in the field as “value alignment,” where developers try to make sure the robot’s core values are aligned with those of humans. One way of doing that, Ravichandar says, is to create a safety mechanism, such as making sure the robot always understands that the best solution it can come up with for a problem might not always be the best answer.

“The time is coming when we will need to have consensus on how to regulate this,” says Ravichandar. “Like any technology, you need to have regulations. But I think it’s absolutely visionary to inject humility into robots, and that’s happening now.”

That’s good news for the rest of us, because killer robots certainly are not the droids we’re looking for.

 

originally written by Colin Poitras

Zhang and His Students Look To Advance Power Systems Into The Future

Peng with the members of his Power and Energy Systems Lab. (UConn Photo/Christopher Larosa)

Over 350 million people in the United States depend on the reliability and consistency of the 450,000 miles of high-voltage lines that form the U.S. power grid to do important daily tasks. With stronger weather events and an increasing number of cyber-attacks, reliable safeguards and technologies are needed to protect this very important utility. The Power and Energy Systems Laboratory, run by Dr. Peng Zhang, F.L. Castleman Associate Professor of Electrical and Computer Engineering and his graduate students, aims at tackling these important issues.

The lab, which focuses on smart grid technology, microgrids, and sustainable energy, has worked on several crucial projects over the past several years, including a dedicated approach to networking the grid system, determining risk assessment models for unintentional islanding of power generators, using ocean waves to generate a sustainable power source, and many other related areas of research.

The lab’s most recent research, which focuses on national infrastructure, is looking to enhance the connectedness of the fractured U.S. grid system:  

“The goal of our research is to make our nation’s energy infrastructure resilient, reliable, secure, and sustainable,” Zhang said. “One of our main areas of focus now is large systems power stability, which is important, because there are no tools available to really assess and predict the status of the system. This tool is highly needed, especially for connected systems.”

That tool, which is being developed from a $1.05 million Department of Energy grant, is being worked on by Zhang and Ph.D. student Yan Li. The idea for the grant was inspired by the learnings Zhang and his students gathered from studying the tools used by local utility company, Eversource Energy:

“The whole U.S. and Canadian grid are connected together, and it’s a huge system, so it’s  therefore very difficult to monitor, assess and control its stability ,” Zhang said. “If you look at Connecticut, companies like Eversource, for the most part, use off-line tools which run many scenarios and only look at snapshots of the system, but that kind of work is not great for analysis, it needs to be monitored and assessed in real-time.”

In particular, Li and Zhang will be forming a formal theory with mathematical rigor, which will be established for computing the bounds of all possible trajectories and estimating the stability margin for the entire system, including the integrated transmission and distribution network.

Furthermore, a new open-source tool via reachable set computations will be developed for real-time dynamic analysis and stability margin calculations. It will be applicable for not only forecasting and monitoring grid performance, but also formally verifying various resiliency enhancement strategies, such as new schemes for system integrity protection and automation to adapt to this evolution of electric networks. 

Zhang and his students are also making significant contributions to knowledge advancement in the state of Connecticut and the region, with their work at the Eversource Energy Center at UConn, which conducts research related to advancing energy technology, as well as performs significant consulting work with Eversource Energy.

Zhang and his students have specifically performed work on research related to unintentional islanding, which is a phenomenon in which a distributed generator continues to be electrified and running, even when the electrical grid surrounding it is no longer active. Traditional methods of detection can often be fooled to think that grid conditions are normal, especially when multiple power generation devices are connected to the same line. 

(UConn Photo/Christopher Larosa)

This kind of scenario is very dangerous to field workers, as the lack of knowledge could cause them to be electrocuted. Zhang said that the research that’s currently being done by himself and his team addresses the creation a risk assessment model to safely avoid danger:

“This kind of research is using machine learning to predict risk in the non-detection zone,” Zhang said. “By non-detection we mean that when an island occurs, there are certain scenarios where the utility company, but with this new risk assessment model, we’ll be able to accurately predict when this scenario is likely to occur.”

Most importantly though, Zhang is happy that the graduate students that he is mentoring, and providing hands-on opportunities to, are getting the necessary experience needed to launch their careers in research and academia.

Ph.D. student Taofeek Orekan, one of the members of the lab, is one of the students that will be using that experience, as he looks for post-grad opportunities in the next few months:

“This is a great lab to start off in,” Orekan said. “I know that any lab that I launch during my career in academia will absolutely be an extension of this lab.”

For more information on the lab, visit http://power.engr.uconn.edu. 

Remembering Charles Knapp, an Engineering Icon

Professor Emeritus Charles Harris Knapp, 86, passed away at

Robert (left) and Charles ’53, ’62 Knapp. (Photo courtesy of The UConn Foundation)

home on Thanksgiving Day

surrounded by his family and loving wife of 62 years, Charleen Gaudet Knapp. Over the course of his 40-year career at UConn, Dr. Knapp enriched and shaped the

lives of innumerable students and inspired an imperishable legacy, The Charles H. Knapp Associate Professorship in Electrical Engineering.

Charlie (or “Red” or “Harris,” as he was variously known) was so eminent in his field, that a handwritten note recommending a student to a graduate program at another well-regarded university signed only “Charlie” was sufficient; no letterhead needed. Colleagues and former students remembering Charlie reiterate his traits as an educator, mentor, and researcher calling him “a true gentleman,” showing people “how to treat others,” “the best teacher,” and never seeing “him lose his cool.”

A lifelong Yankees fan, Charlie was born in New York City, but moved with his family to Coventry, CT at an early age. As an Electrical Engineering undergraduate at UConn, he was the first University Scholar from the School of Engineering and member of Tau Beta Phi, Eta Kappa Nu, and Sigma Xi engineering honor societies. Upon graduation in 1953, Charlie served two years of active duty in the U.S. Air Force before earning his Master’s degree from Yale University and working for RCA and IBM. He returned to UConn in 1958 and became the first candidate awarded a Ph.D. in Electrical Engineering.   

Charlie had many interests outside of academia. He began running in his late 20s, was an avid gardener, active member of the Storrs Congregational Church serving in many capacities including his favorite, singing in the choir for more than 40 years, and spending time with his four children and their families.

In 2012, the Knapp children honored their father with a generous donation establishing The Charles H. Knapp Associate Professorship in Electrical Engineering. Explaining the decision to create an associate professorship, typically given to newer, upcoming faculty, Charlie said, “The younger professors have the freshest education and ideas and are looking for places where they can grow. This professorship will give its holder an edge and will be very helpful in retaining the best associate professors. It is important for the university and the college if we can keep them here, and it’s good for local industry, as well.”

During his lifetime, Professor Knapp enriched the lives of countless students and colleagues, family and friends. Through their lives and his eponymous professorship, his legacy and memory will endure for generations.

 

Donations in his memory may be made to: The Charles H. Knapp Associate Professorship in Electrical Engineering, c/o The UConn Foundation, 2390 Alumni Drive., Unit 3206, Storrs, CT, 06269-3206

Portable Microscope Makes Field Diagnosis Possible

Siddharth Rawat, left, a Ph.D. student, and Bahram Javidi, Board of Trustees Distinguished Professor of Electrical and Computer Engineering, operate a prototype device to examine blood samples for diseases. The portable holographic field microscope offers medical professionals a fast and reliable tool for the identification of diseased cells. (Peter Morenus/UConn Photo)


A portable holographic field microscope developed by UConn optical engineers could provide medical professionals with a fast and reliable new tool for the identification of diseased cells and other biological specimens.

The device, featured in a recent paper published by Applied Optics, uses the latest in digital camera sensor technology, advanced optical engineering, computational algorithms, and statistical analysis to provide rapid automated identification of diseased cells.

One potential field application for the microscope is helping medical workers identify patients with malaria in remote areas of Africa and Asia where the disease is endemic.

Quick and accurate detection of malaria is critical when it comes to treating patients and preventing outbreaks of the mosquito-borne disease, which infected more than 200 million people worldwide in 2015, according to the Centers for Disease Control. Laboratory analysis of a blood sample remains the gold standard for confirming a malaria diagnosis.  Yet access to trained technicians and necessary equipment can be difficult and unreliable in those regions.

The microscope’s potential applications go far beyond the field diagnosis of malaria. The detailed holograms generated by the instrument also can be used in hospitals and other clinical settings for rapid analysis of cell morphology and cell physiology associated with cancer, hepatitis, HIV, sickle cell disease, heart disease, and other illnesses, the developers say.

In checking for the presence of disease, most hospitals currently rely on dedicated laboratories that conduct various tests for cell analysis and identification. But that approach is time consuming, expensive, and labor intensive. It also has to be done by skilled technicians working with the right equipment.

“Our optical instrument cuts down the time it takes to process this information from days to minutes,” says Bahram Javidi, Board of Trustees Distinguished Professor in the Department of Electrical and Computer Engineering and the microscope’s senior developer. “And people running the tests don’t have to be experts, because the algorithms will determine if a result is positive or negative.”

The research team consulted with hematologists, and the algorithms used with the instrument are able to compare a sample against the known features of healthy cells and the known features of diseased cells in order to make proper identification. “It’s all done very quickly,” Javidi says.

How the Device Works

When it comes to identifying patients with malaria, here’s how the device works: A thin smear from a patient’s blood sample is placed on a glass side, which is put under the microscope for analysis. The sample is exposed to a monochromatic light beam generated by a laser diode or other light source. Special components and optical technologies inside the microscope split the light beam into two beams in order to record a digital hologram of the red blood cells in the sample. An image sensor, such as a digital webcam or cell phone camera, connected to the 3-D microscope captures the hologram.  From there, the captured data can be transferred to a laptop computer or offsite laboratory database via the internet. Loaded with dedicated algorithms, the computer or mobile device hardware reconstructs a 3-D profile of the cell and measures the interaction of light with the cell under inspection. Any diseased cells are identified using computer pattern recognition software and statistical analysis.

Quantitative phase profiles of healthy red blood cells (top row) and malaria infected cells (bottom row). (Holographic microscope image courtesy of Bahram Javidi)

Red blood cells infected with the malaria-causing Plasmodium parasite exhibit different properties than healthy blood cells when light passes through them, Javidi says.

“Light behaves differently when it passes through a healthy cell compared to when it passes through a diseased cell,” Javidi says. “Today’s advanced sensors can detect those subtle differences, and it is those nanoscale variations that we are able to measure with this microscope.”

Conventional light microscopes only record the projected image intensity of an object, and have limited capability for visualizing the detailed quantitative characterizations of cells. The digital holograms acquired by UConn’s 3-D microscope, on the other hand, capture unique micro and nanoscale structural features of individual cells with great detail and clarity. Those enhanced images allow medical professionals and researchers to measure an individual cell’s thickness, volume, surface, and dry mass, as well as other structural and physiological changes in a cell or groups of cells over time – all of which can assist in disease identification, treatment, and research. For instance, the device could help researchers see whether new drugs impact cells positively or negatively during clinical trials.

The techniques associated with the holographic microscope also are non-invasive, highlighting its potential use for long-term quantitative analysis of living cells.

Conventional methods of testing blood samples for disease frequently involve labeling, which means the sample is treated with a chemical agent to assist with identification. In the case of malaria, red blood cells are usually treated with a Giemsa stain that reacts to proteins produced by malaria-carrying parasites and thus identifies them. But introducing a chemical into a live cell can change its behavior or damage it.

“If you’re doing an in vitro inspection of stem cells, for instance, and you introduce a chemical agent, you risk damaging those cells. And you can’t do that, because you may want to introduce those cells into the human body at some point,” Javidi says. “Our instrument doesn’t rely on labeling, and therefore avoids that problem.” 

Ph.D. students Tim O’Connor ’17 (ENG), left, Siddharth Rawat, and Adam Markman ’11 (ENG) operate a prototype device to examine blood samples for diseases at the Javidi lab in the Information Technologies Engineering Building. (Peter Morenus/UConn Photo)

The holographic microscope was developed in UConn’s new Multidimensional Optical Sensing & Imaging Systems or MOSIS lab, where Javidi serves as director. The MOSIS lab integrates optics, photonics, and computational algorithms and systems to advance the science and engineering of imaging from nano to macro scales.

A comprehensive report on the MOSIS lab’s work with 3-D optical imaging for medical diagnostics was published last year in Proccedings of the IEEE, the top-ranked journal for electrical and electronics engineering. Joining Javidi in this research are graduate students Adam Markman, Siddharth Rawat, Satoru Komatsu, and Tim O’Connor from UConn; and Arun Anand, an applied optics specialist with Maharaja Sayajirao University of Baroda in Vadodara, India.

The microscope research is supported by Nikon and the National Science Foundation (ECCS 1545687). Students are supported by the U.S. Department of Education, GE, and Canon fellowships. Other sponsors that have supported Javidi’s broader research work and the MOSIS lab over the years include the Defense Advanced Research Projects Agency or DARPA, the U.S. Airforce Research Lab, the U.S. Army, the Office of Naval Research, Samsung, Honeywell, and Lockheed Martin. He has collaborated with colleagues from numerous universities and industries around the world during his time at UConn, including research facilities in Japan, Korea, China, India, Germany, England, Italy, Switzerland, and Spain, among other countries.

Javidi is working with colleagues at UConn Health, including medical oncology and hematology specialist Dr. Biree Andemariam and her staff, for other medical applications. UConn’s tech commercialization office has been involved in discussing potential marketing opportunities for the portable digital microscope. A prototype of the microscope used for initial tests was assembled using 3-D printing technologies, lowering its production costs.

 

Original from UConn Today, Colin Poitras.

Navy Using New UConn Software to Improve Navigation

The Navy is using new software developed by UConn engineering professor Krishna Pattipati to vastly improve the ability to route ships through unpredictable situations.


Major research discoveries generate news headlines. But a research undertaking by one University of Connecticut engineering lab seeks to forestall some headlines of a different kind.

The loss of life because of weather events, as happened on Oct. 1, 2015 when cargo ship El Faro sank with its 33-member crew in Hurricane Joaquin, is one example. Transcripts released by the National Transportation Safety Board showed an increasingly anxious and panicked crew as the 790-foot vessel sailed into the raging storm two years ago.

Software developed by Krishna Pattipati, UTC Professor in Systems Engineering at UConn and his research team, in collaboration with the U.S. Naval Research Laboratory-Monterey, may go a long way toward avoiding such tragedies.

The prototype, named TMPLAR (Tool for Multi-objective Planning and Asset Routing), is now being used by the Navy to vastly improve the ability of ships to reroute through unpredictable weather. It is the type of technology transition that the new National Institute for Undersea Vehicle Technology based at UConn Avery Point, is now able to foster.

Screenshot of a requested ship transit from San Diego, California, toward Alaska. The black line is the suggested route the Navy navigator is given to accept or reject and send on as directions to a ship’s captain. The numbered red circles are ‘waypoints’ along the route, with the starting point labeled ‘0’. These waypoints divide up a possibly long voyage and keep the ship’s path in check.

Created by Pattipati and electrical and computer engineering graduate students David Sidoti, Vinod Avvari, Adam Bienkowski, and Lingyi Zhang, and undergraduate students Matthew Macesker and Michelle Voong, TMPLAR is still in development, but it has already been fully integrated with the Navy’s meteorology and oceanographic weather forecasts.

Members of the UConn team meet weekly with Navy officials, via teleconference, to discuss project updates and receive  feedback.

“Their progress is fast,” says Sidoti. “Frankly, it’s kept us on our toes as we try to manage both our academic responsibilities here at UConn while enhancing and updating the software.”

TMPLAR is like a much more complex version of Google Maps, because it will be applied to ships and submarines, where there is no underlying network of roadways to navigate.

In Google Maps, a user typically seeks to maximize the average speed of travel between start and end locations to get to a destination in the shortest amount of time, hence the route may favor highways instead of back roads.

Pattipati’s team is now approaching problems with upwards of 17 or more objectives, which may change depending on the vehicle and the conditions.

The algorithms take into account obstacles such as ocean depth, undersea pipelines, cables, oil rigs, for example. And they factor in multiple user objectives, whether to traverse to an area to minimize travel time, maximize fuel efficiency given the predicted weather, accomplish training objectives, or maximize operational endurance.

“The tool guarantees safe travel from any point in the ocean, above, on, or below its surface, while making choices en route that optimize fuel consumption and cater to any set of objectives of the operator,” says Sidoti. “Using special clustering techniques, the tool’s algorithms have even been applied to finding low-risk routes that avoid storms or hurricanes.”

The next step for TMPLAR is programming the tool for use by aircraft, such as drones.

Last month, Pattipati and Sidoti traveled to San Diego to demonstrate the capabilities of the software to the Space and Naval Warfare Systems Center Pacific. Their algorithim is now going to be integrated with a tool for aircraft carrier strike group planning.

The lab first published details about the software last year in the journal IEEE, the world’s largest professional organization for the advancement of technology. Avvari, one of the graduate students, will detail some of the enhancements that have been made since then at an upcoming professional conference.

And, as the software transitions to operational settings, the team is looking to speed up the capabilities to output smart weather-informed route recommendations in less than a second. Adding neural network modules to TMPLAR is another new horizon; artificial intelligence would help condense solutions so it is less overwhelming to a user, says Sidoti.

When he reviewed the factors faced by the crew of El Faro using TMPLAR software, Sidoti was able to find safe routes for the ship that involved waiting at waypoints and varying the ship’s speed in order to avoid unsafe environmental conditions, while also reducing costs of the route.

The Coast Guard’s report on the tragedy – released just a month ago – said the captain misjudged the strength of Hurricane Joaquin and should have changed the El Faro’s course.

Sidoti found up to eight possible safe routes using TMPLAR. That’s the sort of information he hopes other captains will have.

Recently, the team received notification that the software was demo’ed to onboard ship navigators who were interested to the point that they requested the ability to use it in order to plan and test it on a real-world deployment.

Funding for this research is supported by the U.S. Office of Naval Research under contracts #N00014-16-1-2036 and #N00014-12-1-0238; by the Naval Research Laboratory under contract #N00173-16-1-G905; and by the Department of Defense High Performance Computing Modernization Program under subproject contract #HPCM034125HQU.

 

Original from UConn Today, Kristen Cole.