Tag Archives: engineering

Microscopic “walkers” find their way across cell surfaces

Technology could provide a way to deliver probes or drugs to cell structures without outside guidance.

By David Chandler


 

CAMBRIDGE, Mass–Nature has developed a wide variety of methods for guiding particular cells, enzymes, and molecules to specific structures inside the body: White blood cells can find their way to the site of an infection, while scar-forming cells migrate to the site of a wound. But finding ways of guiding artificial materials within the body has proven more difficult.

Now a team of researchers at MIT led by Alfredo Alexander-Katz, the Walter Henry Gale Associate Professor of Materials Science and Engineering, has demonstrated a new target-finding mechanism. The new system allows microscopic devices to autonomously find their way to areas of a cell surface, for example, just by detecting an increase in surface friction in places where more cell receptors are concentrated.

The finding is described this week in a paper in the journal Physical Review Letters, written by Alexander-Katz, graduate student Joshua Steimel, and postdoc Juan Aragones.

“The idea was to find out if we could create a synthetic, active system that could sense gradients in biological receptors,” Alexander-Katz explains. “Currently, we don’t know of anything that can do that.”

Cells have a way of locating areas that bear a specific kind of chemical signature — a process called chemotaxis. That’s the method used by white blood cells, for example, to locate regions where pathogens are attacking body cells.

“Our system is very simple,” Alexander-Katz says — similar to the way in which bacteria locate nutrients they need. The system, without guidance, samples areas on a surface and migrates toward those where friction is greater — which also correspond to areas where receptors are concentrated.

The system uses a pair of linked particles with magnetic properties. In the presence of a magnetic field, the paired particles begin to tumble across a surface, with first one particle and then the other making contact — in effect, “walking” across the surface.

So far, the work has been carried out on a model cell surface, on a functionalized microscope slide, but the effect should work similarly with living cells, Alexander-Katz says. The team’s goal now is to demonstrate the ability of the microscopic walkers to find their way toward concentrations of receptors in actual living tissue.

The method could potentially have a variety of applications, Alexander-Katz says. For example, it could be developed as a method of locating tumor cells within the body by identifying their surface texture, perhaps in combination with other characteristics.

Such magnetic microwalkers, he adds, could be unleashed to locate areas of interest on various kinds of surfaces, based solely on differences in friction. The particles naturally migrate toward high-friction regions, where they could then be induced to interact with a surface by active molecules attached to them.

“It’s a very versatile system,” Alexander-Katz says, that can be functionalized by attaching other kinds of receptors or binding agents to affect or monitor the target area in different ways.

The next step is to test the approach in more complex settings. The initial work was done with flat surfaces; the team now aims to conduct studies in complex 3-D settings to make sure the process works effectively in situations that more closely resemble a real cellular environment.

The research was supported by the U.S. Department of Energy, the MIT Energy Initiative, and the Chang family.

Dr Vladlen Shvedov (L) and Dr Cyril Hnatovsky adjusting the hollow laser beam in their lab at RSPE. Image Stuart Hay, ANU

ANU Physicists build reversible tractor beam

We have seen use of laser tractor beams from space ships catching or repelling space ships, objects and people. Science and technology have not developed that much to achieve such feats but Physicists at the Australian National University have done something amazing to push the boundaries science and technology a bit more and closer to that goal.


ANU Laser physicists have built a tractor beam that can repel and attract objects, using a hollow laser beam that is bright around the edges and dark in its centre.

Dr Vladlen Shvedov (L) and Dr Cyril Hnatovsky adjusting the hollow laser beam in their lab at RSPE. Image Stuart Hay, ANU
Dr Vladlen Shvedov (L) and Dr Cyril Hnatovsky adjusting the hollow laser beam in their lab at RSPE. Image Stuart Hay, ANU

It is the first long-distance optical tractor beam and moved particles one fifth of a millimetre in diameter a distance of up to 20 centimetres, around 100 times further than previous experiments.

“Demonstration of a large scale laser beam like this is a kind of holy grail for laser physicists,” said Professor Wieslaw Krolikowski, from the Research School of Physics and Engineering.

The new technique is versatile because it requires only a single laser beam. It could be used, for example, in controlling atmospheric pollution or for the retrieval of tiny, delicate or dangerous particles for sampling.

The researchers can also imagine the effect being scaled up.

“Because lasers retain their beam quality for such long distances, this could work over metres. Our lab just was not big enough to show it,” said co-author Dr Vladlen Shvedov, a driving force behind the ANU project, along with Dr Cyril Hnatovsky.

Unlike previous techniques, which used photon momentum to impart motion, the ANU tractor beam relies on the energy of the laser heating up the particles and the air around them. The ANU team demonstrated the effect on gold-coated hollow glass particles.

The particles are trapped in the dark centre of the beam. Energy from the laser hits the particle and travels across its surface, where it is absorbed creating hotspots on the surface. Air particles colliding with the hotspots heat up and shoot away from the surface, which causes the particle to recoil, in the opposite direction.

To manipulate the particle, the team move the position of the hotspot by carefully controlling the polarisation of the laser beam.

“We have devised a technique that can create unusual states of polarisation in the doughnut shaped laser beam, such as star-shaped (axial) or ring polarised (azimuthal),” Dr Hnatovsky said.

“We can move smoothly from one polarisation to another and thereby stop the particle or reverse its direction at will.”

The work is published in Nature Photonics.

Source : ANU News

The complete electronic sensor, which weighs only 1.8 grams, is imbedded in the cube, and a 3D antenna is positioned around it. Photo credit:  Muhammad Fahad Faroouqi

KAUST research pioneers smart sensors for better and safer living

Atif Shamim and Christian Claudel, KAUST Assistant Professors of Electrical Engineering, work together on creating wireless sensor networks for “smart cities.” It is technology Prof. Shamim describes as “game changing…It will change the way we do many things in our lives, moving us towards smarter living,” he said.

In the “smart cities” of the future, electronic devices and objects will be “smart,” with objects containing sensors that communicate with each other, fixed network nodes and central servers. These sensors are connected through the Internet of things (IOT), which enables them to share information. Intelligent systems at the central servers are then used to analyze and process the data from the sensors.

The complete electronic sensor, which weighs only 1.8 grams, is imbedded in the cube, and a 3D antenna is positioned around it. Photo credit:  Muhammad Fahad Faroouqi
The complete electronic sensor, which weighs only 1.8 grams, is imbedded in the cube, and a 3D antenna is positioned around it. Photo credit: Muhammad Fahad Faroouqi

“The critical component for these processes is low-cost wireless sensing modules,” explained Prof. Shamim. “Fixed sensor nodes are useful, but for these you need a lot of infrastructure, such as towers and assemblies. Our idea is that you would have some fixed sensors, but you would disperse many small, mobile sensors that communicate wirelessly to the fixed sensors, which then communicate all the received information to a central station for analysis.”

COLLABORATING FOR SMART PROGRESS

The use of small, mobile sensors reduces the cost of the infrastructure tremendously, noted Prof. Shamim, and also enables information to be gathered from remote locations where it is difficult or impossible to mount fixed sensors, such as in forests or deserts.

Together, the research groups of Prof. Shamim and Prof. Claudel combined their respective talents and expertise to make progress in using wireless sensors for flood monitoring. This issue is of high importance to Saudi Arabia and cities such as Jeddah, which saw a 2009 catastrophic flood claim the lives of hundreds and cause considerable property damage.

“Classical sensing solutions, such as fixed wireless sensor networks or satellite imagery, are too expensive or too inaccurate to detect floods – and in particular flash floods – well,” noted Prof. Claudel. “Instead, we tested the use of Unmanned Aerial Vehicles (UAVs) equipped with mobile, floatable, 3D printed microsensors and sensor delivery systems to sense and monitor flash flooding events.”

This new system of mobile, floatable sensing, called Lagrangian sensing, “is very promising for large scale sensing, or on-demand sensing, as it requires minimal infrastructure,” the researchers stated. Using this method, UAVs drop the small, disposable wireless sensors over an area to be monitored. The sensors float, so they are carried away by the water flow of the flood. As they move along in the water, they send signals to the UAVs. These signals map the extent of the flood, and this information is transmitted to a central, fixed station for processing. It can then be used to warn the public and other authorities about the extent of the flood.

“Prof. Claudel carries out the systems level design and implementation for the research project, and my group develops the actual physical sensors,” said Prof. Shamim. “In that way, I believe we are a very good fit for collaboration.”

Their collaboration produced a paper recently published in IEEE Transactions on Antennas & Propagation, entitled “An Inkjet-Printed Buoyant 3-D Lagrangian Sensor for Real-Time Flood Monitoring” (DOI: 10.1109/TAP.2014.2309957). KAUST has applied for patent protection for this and other related technologies.

DEVELOPING LOW-COST SOLUTIONS

One of the challenges Profs. Claudel and Shamim and their teams faced in the research work was designing the sensors. “We wanted to make them low-cost so they are basically disposable,” explained Prof. Shamim. “We use inkjet printers to print electronics on paper and plastics, but in this case we used paper, as it is lightweight, 1/10th the cost of plastic, and is very suitable for inkjet printing. In addition, it is biodegradable and comes from a renewable resource.”

The researchers produced a small paper cube with a size of 13 mm x 13 mm x 13 mm. The complete electronic sensor, which weighs only 1.8 grams, is imbedded in the cube, and a 3D antenna is positioned around it, enabling the cube to give a signal in any direction it is moving (or floating).

“Because we were working on a flood monitoring application, we had to optimize the sensor to work in water as well as in air,” Prof. Shamim noted. “We were skeptical about its performance in water, so we sealed it with a special glue. We then produced a cube that is very small, lightweight, floats in water, and is electrically sealed. It works very well in water and radiates up to 50 meters in all directions. The performance was better than we expected.”

The technology has many other possible applications: “You could integrate sensors for ammonia, sulfur, carbon monoxide, humidity, or temperature into the cube,” said Prof. Shamim. “This would allow for detection of poisonous gases and other environmental conditions, which would be especially helpful in industrial settings and in remote locations, such as during forest fire events.”

Profs. Shamim and Claudel want to integrate their low-cost, printable, and disposable microsensor technology into the day-to-day lives of everyday people. Not only would the technology enable greater safety for individuals during catastrophic events such as floods, but it could also assist in locating cars in busy parking lots, tracking expired foods in supermarkets, and in creating smart houses, where household appliances and electronic lock systems “talk” to each other to make sure they are in proper working order.

“I believe this technology will change the way people live, shop, and monitor things,” said Prof. Shamim. “We will have better living, from our homes to our offices to our industries – and that is a benefit for all.”

Sourse: KAUST

The NOMADD technology represents KAUST's first royalty-bearing license agreement. Credit: KAUST News

Innovation in the desert! KAUST’s NOMADD sets sights on solar energy future

The NOMADD technology represents KAUST’s first royalty-bearing license agreement.

By Meres J. Weche


The United Nations estimates the Saudi population will grow to 45 million by 2050; and as the population increases, domestic energy demand is anticipated to double by 2030. In recognition of the growing importance of developing sustainable and renewable energy sources for the Kingdom, the Saudi government has established the ambitious goal of generating a third of the country’s electricity sources (41,000 megawatts) through solar power by 2032. Towards this goal, the King Abdullah City for Atomic and Renewable Energy (KACARE) aims to construct a $109 billion solar industry in Saudi Arabia, which would represent about 20,000 football fields worth of solar panels.

“We hope to be the industry standard solution to clean all those panels,” said Georg Eitelhuber, Founder and Chief Executive Officer of NOMADD. The startup company, developed three years ago at KAUST and originally supported and funded by theEntrepreneurship Center and the Seed Fund program, offers a waterless and remotely operated system to clean solar panels. The acronym NOMADD stands for NO-water Mechanical Automated Dusting Device.

The NOMADD technology represents KAUST's first royalty-bearing license agreement. Credit: KAUST News
The NOMADD technology represents KAUST’s first royalty-bearing license agreement. Credit: KAUST News

Describing the challenges facing Saudi Arabia’s burgeoning solar energy industry, the NOMADD founder says: “The big challenge, is dust. Desert winds pick up the dust and push it onto the solar panels, all day every day. Sometimes you can have dust storms which put so much dust on the solar panel surface, you can lose 60% of your output in a single day.” Actually, solar panels lose between 0.4-0.8% of their efficiency per day just from desert sand and dust.

A mechanical engineer by training, Eitelhuber was working as a physics teacher at the KAUST School when he started experimenting with Lego blocks and paper to find a solution to clean solar panels exposed to the rough dusty environment of Saudi Arabia. His innovation has since been recognized with the 2014 Solar Pioneer Award and he has been working on further testing and developing the solution with world-leading companies in solar energy such as First Solar Inc. and SunPower Corp.

Eitelhuber is grateful for the backing of KAUST, with all of its resources, in assisting inventors like himself. As the NOMADD team works with various industrial testing partners on improving the technology, KAUST Tech Transfer is there to maintain control of patentable technology which may emerge in the process. A milestone was achieved last month when KAUST signed its first royalty-bearing license agreement for the NOMADD desert solar solution system.

A Continuous Drive for Improvement

Demonstrating the newly devised fifth version of the NOMADD system in its three years of development, Georg Eitelhuber explains that it’s now “70% lighter than previous versions and uses less than half of the power.” In addition to that, it’s much cheaper to manufacture.

“Every time we do a new version it’s simpler, cheaper and faster,” he adds. For example, the rail system supporting the brushes cleaning the solar panels from top to bottom is not only lighter and cheaper but it also now just clips on – whereas previous versions required many nuts and bolts. The mounting system moreover features an inbuilt self-adjustment process tailored to determine the optimal gravity-adjusted angle as the solar panels are cleaned.

It’s important for the cleaning system to be both economically and functionally optimized since some panel rows can be 400 meters long. “That’s a lot of rail,” said Eitelhuber.” “The old version had literally hundreds of nuts and bolts, little fasteners and washers and it worked great but it also weighted as much as a tank.”

Compared to some earlier models, which had around 120-odd manufacturing pieces, the latest NOMADD system has narrowed it down to 10 to 15 pieces. This means that it’s now easier to manufacture and assemble. “The key thing is that it has to be cheaper than sending out a worker with a squeegee and more economical than anything else in the market,” he adds.

The achieved objective has been to make NOMADD desert-proof – as the arid environment causes things to break down at higher frequencies. The device is basically machined aluminum and stainless steel.

It’s also noteworthy that the brushes used to non-abrasively clean the solar panels can easily be slid out and replaced. So it would take someone around five minutes to change all the brushes.

In addition, one of the major advantages of the NOMADD system is that it’s remotely operated. The cleaning functions can be monitored and operated online from around the world.

A Saudi-Specific Innovation with a Global Footprint

“The advantage that we’ve got is that we’ve basically been three years in development and we’ve been developing this solution for the desert while being in the desert. We’ve got a real understanding of the issues involved in cleaning solar panels in the desert,” said Georg Eitelhuber.

Unlike some other solar panel cleaning solutions from North American and European companies, designed for mild climates, that use water and require manual labor, the NOMADD system really has an edge by being a waterless model ideally suited for these arid conditions. “We understand that having someone standing outside at 45 degrees Celsius cleaning solar panels eight hours a day isn’t feasible,” he adds.

As they keep an eye out for the competition, the NOMADD team is confident that, once they make it through the final development process, they will have every chance of being a huge commercial success.

KAUST’s director of New Ventures and Entrepreneurship, Gordon McConnell, says NOMADD’s local presence in the Kingdom will help contribute in building a knowledge-based economy in Saudi Arabia. “The local incorporation is not just of bureaucratic significance, but will now enable NOMADD to develop its business which in turn will help to create high level jobs in sales, marketing and technical areas, while also offering an opportunity to build up local manufacturing capacity and it will make it easier for fund raising within the Kingdom,” said McConnell.

The NOMADD project has greatly benefited from the collaborative efforts of several key team members such as Guodong Li, Chief Electrical Engineer, and Elizabeth Cassell, the project’s chief Administrator, both from the KAUST Solar Center; as well as Head Mechanical Design Engineer Steven Schneider who has been instrumental in producing technical drawings for manufacturing. Andres Pablo, a Ph.D. student, and Razeen Stoffberg, one of Georg’s ex students front he KAUST school, have been assisting with technical setups and product testing and evaluation.

Also, as much of the manufacturing work is done in Asia, the NOMADD team has set up an office in Singapore, headed by Chief Development Officer Cliff Barrett. As a next step, the team has been actively recruiting a new CEO to help the project achieve critical mass and reach their ambitious future milestones.

“Thanks to some great mentorship from the KAUST New Ventures and Entrepreneurshipteam, I’ve done my best as a CEO but I’m an engineer and an inventor by nature,” said Georg Eitelhuber. “It’s been one of my dreams from the very beginning to try and start something which will have a net positive environmental and social impact.”

Source: KAUST News

How to hide like an octopus : Researchers create materials that reproduce cephalopods’ ability to quickly change colors and textures

By David Chandler


CAMBRIDGE, Mass– Cephalopods, which include octopuses, squid, and cuttlefish, are among nature’s most skillful camouflage artists, able to change both the color and texture of their skin within seconds to blend into their surroundings — a capability that engineers have long struggled to duplicate in synthetic materials. Now a team of researchers has come closer than ever to achieving that goal, creating a flexible material that can change its color or fluorescence and its texture at the same time, on demand, by remote control.

The results of their research have been published in the journal Nature Communications, in a paper by a team led by MIT Assistant Professor of Mechanical Engineering Xuanhe Zhao and Duke University Professor of Chemistry Stephen Craig.

Zhao, who joined the MIT faculty from Duke this month and holds a joint appointment with the Department of Civil and Environmental Engineering, says the new material is essentially a layer of electro-active elastomer that could be quite easily adapted to standard manufacturing processes and uses readily available materials. This could make it a more economical dynamic camouflage material than others that are assembled from individually manufactured electronic modules.

While its most immediate applications are likely to be military, Zhao says the same basic approach could eventually lead to production of large, flexible display screens and anti-fouling coatings for ships.

In its initial proof-of-concept demonstrations, the material can be configured to respond with changes in both texture and fluorescence, or texture and color. In addition, while the present version can produce a limited range of colors, there is no reason that the range of the palette cannot be increased, Craig says.

Learning from nature

Cephalopods achieve their remarkable color changes using muscles that can alter the shapes of tiny pigment sacs within the skin — for example, contracting to change a barely visible round blob of color into a wide, flattened shape that is clearly seen. “In a relaxed state, it is very small,” Zhao says, but when the muscles contract, “they stretch that ball into a pancake, and use that to change color. The muscle contraction also varies skin textures, for example, from smooth to bumpy.” Octopuses use this mechanism both for camouflage and for signaling, he says, adding, “We got inspired by this idea, from this wonderful creature.”

The new synthetic material is a form of elastomer, a flexible, stretchable polymer. “It changes its fluorescence and texture together, in response to a change in voltage applied to it — essentially, changing at the flip of a switch,” says Qiming Wang, an MIT postdoc and the first author of the paper.

“We harnessed a physical phenomenon that we discovered in 2011, that applying voltage can dynamically change surface textures of elastomers,” Zhao says.

“The texturing and deformation of the elastomer further activates special mechanically responsive molecules embedded in the elastomer, which causes it to fluoresce or change color in response to voltage changes,” Craig adds. “Once you release the voltage, both the elastomer and the molecules return to their relaxed state — like the cephalopod skin with muscles relaxed.”

Multiple uses for quick changes

While troops and vehicles often move from one environment to another, they are presently limited to fixed camouflage patterns that might be effective in one environment but stick out like a sore thumb in another. Using a system like this new elastomer, Zhao suggests, either on uniforms or on vehicles, could allow the camouflage patterns to constantly change in response to the surroundings.

“The U.S. military spends millions developing different kinds of camouflage patterns, but they are all static,” Zhao says. “Modern warfare requires troops to deploy in many different environments during single missions. This system could potentially allow dynamic camouflage in different environments.”

Another important potential application, Zhao says, is for an anti-fouling coating on the hulls of ships, where microbes and creatures such as barnacles can accumulate and significantly degrade the efficiency of the ship’s propulsion. Earlier experiments have shown that even a brief change in the surface texture, from the smooth surface needed for fast movement to a rough, bumpy texture, can quickly remove more than 90 percent of the biological fouling.

In addition to Zhao, Craig, and Wang, the team also included Duke student Gregory Grossweiler. The work was supported by the U.S. Office of Naval Research, the U.S. Army Research Laboratory and Army Research Office, and the National Science Foundation.

Source : MIT News Office

The power of salt

MIT study investigates power generation from the meeting of river water and seawater.

By Jennifer Chu


Where the river meets the sea, there is the potential to harness a significant amount of renewable energy, according to a team of mechanical engineers at MIT.

The researchers evaluated an emerging method of power generation called pressure retarded osmosis (PRO), in which two streams of different salinity are mixed to produce energy. In principle, a PRO system would take in river water and seawater on either side of a semi-permeable membrane. Through osmosis, water from the less-salty stream would cross the membrane to a pre-pressurized saltier side, creating a flow that can be sent through a turbine to recover power.

The MIT team has now developed a model to evaluate the performance and optimal dimensions of large PRO systems. In general, the researchers found that the larger a system’s membrane, the more power can be produced — but only up to a point. Interestingly, 95 percent of a system’s maximum power output can be generated using only half or less of the maximum membrane area.

Leonardo Banchik, a graduate student in MIT’s Department of Mechanical Engineering, says reducing the size of the membrane needed to generate power would, in turn, lower much of the upfront cost of building a PRO plant.

“People have been trying to figure out whether these systems would be viable at the intersection between the river and the sea,” Banchik says. “You can save money if you identify the membrane area beyond which there are rapidly diminishing returns.”

Banchik and his colleagues were also able to estimate the maximum amount of power produced, given the salt concentrations of two streams: The greater the ratio of salinities, the more power can be generated. For example, they found that a mix of brine, a byproduct of desalination, and treated wastewater can produce twice as much power as a combination of seawater and river water.

Based on his calculations, Banchik says that a PRO system could potentially power a coastal wastewater-treatment plant by taking in seawater and combining it with treated wastewater to produce renewable energy.

“Here in Boston Harbor, at the Deer Island Waste Water Treatment Plant, where wastewater meets the sea … PRO could theoretically supply all of the power required for treatment,” Banchik says.

He and John Lienhard, the Abdul Latif Jameel Professor of Water and Food at MIT, along with Mostafa Sharqawy of King Fahd University of Petroleum and Minerals in Saudi Arabia, report their results in the Journal of Membrane Science.

Finding equilibrium in nature

The team based its model on a simplified PRO system in which a large semi-permeable membrane divides a long rectangular tank. One side of the tank takes in pressurized salty seawater, while the other side takes in river water or wastewater. Through osmosis, the membrane lets through water, but not salt. As a result, freshwater is drawn through the membrane to balance the saltier side.

“Nature wants to find an equilibrium between these two streams,” Banchik explains.

As the freshwater enters the saltier side, it becomes pressurized while increasing the flow rate of the stream on the salty side of the membrane. This pressurized mixture exits the tank, and a turbine recovers energy from this flow.

Banchik says that while others have modeled the power potential of PRO systems, these models are mostly valid for laboratory-scale systems that incorporate “coupon-sized” membranes. Such models assume that the salinity and flow of incoming streams is constant along a membrane. Given such stable conditions, these models predict a linear relationship: the bigger the membrane, the more power generated.

But in flowing through a system as large as a power plant, Banchik says, the streams’ salinity and flux will naturally change. To account for this variability, he and his colleagues developed a model based on an analogy with heat exchangers.

“Just as the radiator in your car exchanges heat between the air and a coolant, this system exchanges mass, or water, across a membrane,” Banchik says. “There’s a method in literature used for sizing heat exchangers, and we borrowed from that idea.”

The researchers came up with a model with which they could analyze a wide range of values for membrane size, permeability, and flow rate. With this model, they observed a nonlinear relationship between power and membrane size for large systems. Instead, as the area of a membrane increases, the power generated increases to a point, after which it gradually levels off. While a system may be able to produce the maximum amount of power at a certain membrane size, it could also produce 95 percent of the power with a membrane half as large.

Still, if PRO systems were to supply power to Boston’s Deer Island treatment plant, the size of a plant’s membrane would be substantial — at least 2.5 million square meters, which Banchik notes is the membrane area of the largest operating reverse osmosis plant in the world.

“Even though this seems like a lot, clever people are figuring out how to pack a lot of membrane into a small volume,” Banchik says. “For example, some configurations are spiral-wound, with flat sheets rolled up like paper towels around a central tube. It’s still an active area of research to figure out what the modules would look like.”

“Say we’re in a place that could really use desalinated water, like California, which is going through a terrible drought,” Banchik adds. “They’re building a desalination plant that would sit right at the sea, which would take in seawater and give Californians water to drink. It would also produce a saltier brine, which you could mix with wastewater to produce power. More research needs to be done to see whether it can be economically viable, but the science is sound.”

This work was funded by the King Fahd University of Petroleum and Minerals through the Center for Clean Water and Clean Energy and by the National Science Foundation.

Source: MIT News Office

Amazon's delivery drones. Credit: Amaon

Delivery by drone

New algorithm lets drones monitor their own health during long package-delivery missions.

By Jennifer  Chu


CAMBRIDGE, MA — In the near future, the package that you ordered online may be deposited at your doorstep by a drone: Last December, online retailer Amazon announced plans to explore drone-based delivery, suggesting that fleets of flying robots might serve as autonomous messengers that shuttle packages to customers within 30 minutes of an order.

To ensure safe, timely, and accurate delivery, drones would need to deal with a degree of uncertainty in responding to factors such as high winds, sensor measurement errors, or drops in fuel. But such “what-if” planning typically requires massive computation, which can be difficult to perform on the fly.

Now MIT researchers have come up with a two-pronged approach that significantly reduces the computation associated with lengthy delivery missions. The team first developed an algorithm that enables a drone to monitor aspects of its “health” in real time. With the algorithm, a drone can predict its fuel level and the condition of its propellers, cameras, and other sensors throughout a mission, and take proactive measures — for example, rerouting to a charging station — if needed.

The researchers also devised a method for a drone to efficiently compute its possible future locations offline, before it takes off. The method simplifies all potential routes a drone may take to reach a destination without colliding with obstacles.

In simulations involving multiple deliveries under various environmental conditions, the researchers found that their drones delivered as many packages as those that lacked health-monitoring algorithms — but with far fewer failures or breakdowns.

Amazon's delivery drones. Credit: Amaon
Amazon’s delivery drones. Credit: Amaon

“With something like package delivery, which needs to be done persistently over hours, you need to take into account the health of the system,” says Ali-akbar Agha-mohammadi, a postdoc in MIT’s Department of Aeronautics and Astronautics. “Interestingly, in our simulations, we found that, even in harsh environments, out of 100 drones, we only had a few failures.”

Agha-mohammadi will present details of the group’s approach in September at the IEEE/RSJ International Conference on Intelligent Robots and Systems, in Chicago. His co-authors are MIT graduate student Kemal Ure; Jonathan How, the Richard Cockburn Maclaurin Professor of Aeronautics and Astronautics; and John Vian of Boeing.

Tree of possibilities

Planning an autonomous vehicle’s course often involves an approach called Markov Decision Process (MDP), a sequential decision-making framework that resembles a “tree” of possible actions. Each node along a tree can branch into several potential actions — each of which, if taken, may result in even more possibilities. As Agha-mohammadi explains it, MDP is “the process of reasoning about the future” to determine the best sequence of policies to minimize risk.

MDP, he says, works reasonably well in environments with perfect measurements, where the result of one action will be observed perfectly. But in real-life scenarios, where there is uncertainty in measurements, such sequential reasoning is less reliable. For example, even if a command is given to turn 90 degrees, a strong wind may prevent that command from being carried out.

Instead, the researchers chose to work with a more general framework of Partially Observable Markov Decision Processes (POMDP). This approach generates a similar tree of possibilities, although each node represents a probability distribution, or the likelihood of a given outcome. Planning a vehicle’s route over any length of time, therefore, can result in an exponential growth of probable outcomes, which can be a monumental task in computing.

Agha-mohammadi chose to simplify the problem by splitting the computation into two parts: vehicle-level planning, such as a vehicle’s location at any given time; and mission-level, or health planning, such as the condition of a vehicle’s propellers, cameras, and fuel levels.

For vehicle-level planning, he developed a computational approach to POMDP that essentially funnels multiple possible outcomes into a few most-likely outcomes.

“Imagine a huge tree of possibilities, and a large chunk of leaves collapses to one leaf, and you end up with maybe 10 leaves instead of a million leaves,” Agha-mohammadi says. “Then you can … let this run offline for say, half an hour, and map a large environment, and accurately predict the collision and failure probabilities on different routes.”

He says that planning out a vehicle’s possible positions ahead of time frees up a significant amount of computational energy, which can then be spent on mission-level planning in real time. In this regard, he and his colleagues used POMDP to generate a tree of possible health outcomes, including fuel levels and the status of sensors and propellers.

Proactive delivery

The researchers combined the two computational approaches, and ran simulations in which drones were tasked with delivering multiple packages to different addresses under various wind conditions and with limited fuel. They found that drones operating under the two-pronged approach were more proactive in preserving their health, rerouting to a recharge station midmission to keep from running out of fuel. Even with these interruptions, the team found that these drones were able to deliver just as many packages as those that were programmed to simply make deliveries without considering health.

Going forward, the team plans to test the route-planning approach in actual experiments. The researchers have attached electromagnets to small drones, or quadrotors, enabling them to pick up and drop off small parcels. The team has also programmed the drones to land on custom-engineered recharge stations.

“We believe in the near future, in a lab setting, we can show what we’re gaining with this framework by delivering as many packages as we can while preserving health,” Agha-mohammadi says. “Not only the drone, but the package might be important, and if you fail, it could be a big loss.”

This work was supported by Boeing.

Source: MIT News office

Imagine a city that thinks about your safety

Can a city be smart? The scientists and technology experts at KAUST certainly think so. They have been working on a number of smart solutions to help deal with issues like traffic congestion, water management, and urban flooding.

Raghid Shreih, a Technology Portfolio Manager at KAUST’s Technology, Transfer and Innovation Division (TTI), works with KAUST researchers to protect, manage and commercialize KAUST’s intellectual property portfolio. He’s been involved with evaluating many of the smart city systems developed at the University.

“The world’s urban population is growing very rapidly,” says Shreih, “and this is presenting a lot of new challenges for cities, particularly in terms of urban planning and infrastructure, public transit, traffic congestion and pollution. As cities become more densely populated, there is also the risk of severe weather incidents causing a lot of damage and casualties. KAUST researchers are developing solutions to address some of these problems.”

KAUST Video:Flood and Traffic Monitoring System

Our latest technology video explains the integrated sensor system for monitoring urban floods and traffic congestions.

One of these solutions is a dual-usage wireless sensor system that tracks traffic congestion and flood incidents in cities. Using a combination of ultrasonic range finders and infrared thermal sensors, the system can monitor traffic flow and roadway flooding, and can be deployed on a large urban scale to provide real-time, highly accurate data on current conditions.

“Because flash floods are extremely rare events, there is not really an incentive to deploy a dedicated infrastructure to address these problems,” said Prof. Christian Claudel, lead inventor of the system. “We wanted to have traffic sensors that would also be capable of detecting flash floods as a secondary application—therefore, the marginal cost of sensing flash floods is zero.”

With storms and floods accounting for nearly 70 percent of the world’s natural disasters, this smart technology can provide up-to-the-minute warnings and allow rapid response to emergency situations. The data collected from these sensors is sent to central servers for assimilation with satellite data, forming real-time maps and forecasting the future path, intensity, and speed of floods and traffic.

This is just one of the many smart systems developed at KAUST.

Shreih says, “The technology being developed at KAUST can be adopted by company and industry partners who will be able to integrate it within their systems and use it to build new infrastructure projects for the cities of the future.”

Source: KAUST

Collecting just the right data : MIT Research

When you can’t collect all the data you need, a new algorithm tells you which to target.

 Larry Hardesty | MIT News Office 


Much artificial-intelligence research addresses the problem of making predictions based on large data sets. An obvious example is the recommendation engines at retail sites like Amazon and Netflix.

But some types of data are harder to collect than online click histories —information about geological formations thousands of feet underground, for instance. And in other applications — such as trying to predict the path of a storm — there may just not be enough time to crunch all the available data.

Dan Levine, an MIT graduate student in aeronautics and astronautics, and his advisor, Jonathan How, the Richard Cockburn Maclaurin Professor of Aeronautics and Astronautics, have developed a new technique that could help with both problems. For a range of common applications in which data is either difficult to collect or too time-consuming to process, the technique can identify the subset of data items that will yield the most reliable predictions. So geologists trying to assess the extent of underground petroleum deposits, or meteorologists trying to forecast the weather, can make do with just a few, targeted measurements, saving time and money.

Levine and How, who presented their work at the Uncertainty in Artificial Intelligence conference this week, consider the special case in which something about the relationships between data items is known in advance. Weather prediction provides an intuitive example: Measurements of temperature, pressure, and wind velocity at one location tend to be good indicators of measurements at adjacent locations, or of measurements at the same location a short time later, but the correlation grows weaker the farther out you move either geographically or chronologically.

Graphic content

Such correlations can be represented by something called a probabilistic graphical model. In this context, a graph is a mathematical abstraction consisting of nodes — typically depicted as circles — and edges — typically depicted as line segments connecting nodes. A network diagram is one example of a graph; a family tree is another. In a probabilistic graphical model, the nodes represent variables, and the edges represent the strength of the correlations between them.

Levine and How developed an algorithm that can efficiently calculate just how much information any node in the graph gives you about any other — what in information theory is called “mutual information.” As Levine explains, one of the obstacles to performing that calculation efficiently is the presence of “loops” in the graph, or nodes that are connected by more than one path.

Calculating mutual information between nodes, Levine says, is kind of like injecting blue dye into one of them and then measuring the concentration of blue at the other. “It’s typically going to fall off as we go further out in the graph,” Levine says. “If there’s a unique path between them, then we can compute it pretty easily, because we know what path the blue dye will take. But if there are loops in the graph, then it’s harder for us to compute how blue other nodes are because there are many different paths.”

So the first step in the researchers’ technique is to calculate “spanning trees” for the graph. A tree is just a graph with no loops: In a family tree, for instance, a loop might mean that someone was both parent and sibling to the same person. A spanning tree is a tree that touches all of a graph’s nodes but dispenses with the edges that create loops.

Betting the spread

Most of the nodes that remain in the graph, however, are “nuisances,” meaning that they don’t contain much useful information about the node of interest. The key to Levine and How’s technique is a way to use those nodes to navigate the graph without letting their short-range influence distort the long-range calculation of mutual information.

That’s possible, Levine explains, because the probabilities represented by the graph are Gaussian, meaning that they follow the bell curve familiar as the model of, for instance, the dispersion of characteristics in a population. A Gaussian distribution is exhaustively characterized by just two measurements: the average value — say, the average height in a population — and the variance — the rate at which the bell spreads out.

“The uncertainty in the problem is really a function of the spread of the distribution,” Levine says. “It doesn’t really depend on where the distribution is centered in space.” As a consequence, it’s often possible to calculate variance across a probabilistic graphical model without relying on the specific values of the nodes. “The usefulness of data can be assessed before the data itself becomes available,” Levine says.

Reprinted with permission of MIT News (http://newsoffice.mit.edu/)