Mediaboss Marketing

DENVER With wildfires becoming bigger and more destructive as the West dries out and heats up, agencies and officials tasked with preventing and battling the blazes could soon have a new tool to add to their arsenal of prescribed burns, pick axes, chain saws and aircraft.

The high-tech help could come by way of an area not normally associated with fighting wildfires: artificial intelligence. And space.

Lockheed Martin Space, based in Jefferson County, is tapping decades of experience of managing satellites, exploring space and providing information for the U.S. military to offer more accurate data quicker to ground crews. They are talking to the U.S. Forest Service, university researchers and a Colorado state agency about how their their technology could help.

By generating more timely information about on-the-ground conditions and running computer programs to process massive amounts of data, Lockheed Martin representatives say they can map fire perimeters in minutes rather than the hours it can take now. They say the artificial intelligence, or AI, and machine learning the company has applied to military use can enhance predictions about a fires direction and speed.

The scenario that wildland fire operators and commanders work in is very similar to that of the organizations and folks who defend our homeland and allies. Its a dynamic environment across multiple activities and responsibilities, said Dan Lordan, senior manager for AI integration at Lockheed Martins Artificial Intelligence Center.

Lockheed Martin aims to use its technology developed over years in other areas to reduce the time it takes to gather information and make decisions about wildfires, said Rich Carter, business development director for Lockheed Martin Spaces Mission Solutions.

The quicker you can react, hopefully then you can contain the fire faster and protect peoples properties and lives, Carter said.

The concept of a regular fire season has all but vanished as drought and warmer temperatures make Western lands ripe for ignition. At the end of December, the Marshall fire burned 991 homes and killed two people in Boulder County. The Denver area just experienced its third driest-ever April with only 0.06 of an inch of moisture, according to the National Weather Service.

Colorado had the highest number of fire-weather alerts in April than any other April in the past 15 years. Crews have quickly contained wind-driven fires that forced evacuations along the Front Range and on the Eastern Plains. But six families in Monte Vista lost their homes in April when a fire burned part of the southern Colorado town.

Since 2014, the Colorado Division of Fire Prevention and Control has flown planes equipped with infrared and color sensors to detect wildfires and provide the most up-to-date information possible to crews on the ground. The onboard equipment is integrated with the Colorado Wildfire Information System, a database that provides images and details to local fire managers.

Last year we found almost 200 new fires that nobody knew anything about, said Bruce Dikken, unit chief for the agencys multi-mission aircraft program. I dont know if any of those 200 fires would have become big fires. I know they didnt become big fires because we found them.

When the two Pilatus PC-12 airplanes began flying in 2014, Colorado was the only state with such a program conveying the information in near real time, Dikken said. Lockheed Martin representatives have spent time in the air on the planes recently to see if its AI can speed up the process.

We dont find every single fire that we fly over and it can certainly be faster if we could employ some kind of technology that might, for instance, automatically draw the fire perimeter, Dikken said. Right now, its very much a manual process.

Something like the 2020 Cameron Peak fire, which at 208,663 acres is Colorados largest wildfire, could take hours to map, Dikken said.

And often the people on the planes are tracking several fires at the same time. Dikken said the faster they can collect and process the data on a fires perimeter, the faster they can move to the next fire. If it takes a couple of hours to map a fire, what I drew at the beginning may be a little bit different now, he said.

Lordan said Lockheed Martin engineers who have flown with the state crews, using the video and images gathered on the flights, have been able to produce fire maps in as little as 15 minutes.

The company has talked to the state about possibly carrying an additional computer that could help crunch all that information and transmit the map of the fire while still in flight to crews on the ground, Dikken said. The agency is waiting to hear the results of Lockheed Martins experiences aboard the aircraft and how the AI might help the state, he added.

Actionable intelligence

The company is also talking to researchers at the U.S. Forest Service Missoula Fire Sciences Laboratory in Montana. Mark Finney, a research forester, said its early in discussions with Lockheed Martin.

They have a strong interest in applying their skills and capabilities to the wildland fire problem, and I think that would be welcome, Finney said.

The lab in Missoula has been involved in fire research since 1960 and developed most of the fire-management tools used for operations and planning, Finney said. Were pretty well situated to understand where new things and capabilities might be of use in the future and some of these things certainly might be.

However, Lockheed Martin is focused on technology and thats not really been where the most effective use of our efforts would be, Finney said.

Prevention and mitigation and preemptive kind of management activities are where the great opportunities are to change the trajectory were on, Finney said. Improving reactive management is unlikely to yield huge benefits because the underlying source of the problem is the fuel structure across large landscapes as well as climate change.

Logging and prescribed burns, or fires started under controlled conditions, are some of the management practices used to get rid of fuel sources or create a more diverse landscape. But those methods have sometimes met resistance, Finney said.

As bad as the Cameron Peak fire was, Finney said the prescribed burns the Arapaho and Roosevelt National Forests did through the years blunted the blazes intensity and changed the flames movement in spots.

Unfortunately, they hadnt had time to finish their planned work, Finney said.

Lordan said the value of artificial intelligence, whether in preventing fires or responding to a fire, is producing accurate and timely information for fire managers, what he called actionable intelligence.

One example, Lordan said, is information gathered and managed by federal agencies on the types and conditions of vegetation across the country. He said updates are done every two to three two years. Lockheed Martin uses data from satellites managed by the European Space Agency that updates the information about every five days.

Lockheed is working with Nvidia, a California software company, to produce a digital simulation of a wildfire based on an areas topography, condition of the vegetation, wind and weather to help forecast where and how it will burn. After the fact, the companies used the information about the Cameron Peak fire, plugging in the more timely satellite data on fuel conditions, and generated a video simulation that Lordan said was similar to the actual fires behavior and movement.

While appreciating the help technology provides, both Dikken with the state of Colorado and Finney with the Forest Service said there will always be a need for ground-truthing by people.

Applying AI to fighting wildfires isnt about taking people out of the loop, Lockheed Martin spokesman Chip Eschenfelder said. Somebody will always be in the loop, but people currently in the loop are besieged by so much data they cant sort through it fast enough. Thats where this is coming from.

Invalid username/password.

Please check your email to confirm and complete your registration.

Use the form below to reset your password. When you've submitted your account email, we will send an email with a reset code.

Previous

Next

See the rest here:
Artificial intelligence tapped to fight Western wildfires - Portland Press Herald - Press Herald

BIRMINGHAM, United Kingdom Could artificial intelligence finally make your morning commute smooth and relatively traffic-free? Researchers from Aston University report that their new AI traffic light system effectively keeps the flow of traffic rolling and mitigates congestion by reading live camera footage and adapting traffic lights on the fly.

Simply put, if theres no cars coming from the other direction, say goodbye to those long red lights clogging up the street!

The AI utilizes a type of learning called deep reinforcement, which means the program understands when it isnt doing well (traffic is bad) and reacts. As time goes on, the algorithm learns more and more based on better results.

During a round of assessments, this first-of-its-kind AI outperformed all other tested methods. The other methods relied mostly on manually-designed phase transitions.

The research team developed and constructed a cutting-edge, photo-realistic traffic simulator called Traffic 3Dto train the AI. Traffic 3D taught the program how to best react to various traffic and weather scenarios.

The AI was then tested on real junction footage. Sure enough, it adapted well to real traffic intersections despite being trained entirely on simulations up until that point. Study authors say this indicates the AI would be effective across many real-world settings.

We have set this up as a traffic control game. The program gets a reward when it gets a car through a junction. Every time a car has to wait or theres a jam, theres a negative reward. Theres actually no input from us; we simply control the reward system, says Dr. Maria Chli, a reader in Computer Science, in a university release.

Today, most traffic light automation systems at junctions rely on magnetic induction loops,or a wire that sits on the road and recognizes when cars pass over it. The program then reacts to that stimuli. This newly devised AI, however, is able to see high traffic volume before cars have even passed the lights. It is much more responsive and can react more quickly.

The reason we have based this program on learned behaviors is so that it can understand situations it hasnt explicitly experienced before. Weve tested this with a physical obstacle that is causing congestion, rather than traffic light phasing, and the system still did well. As long as there is a causal link, the computer will ultimately figure out what that link is. Its an intensely powerful system, explains Dr. George Vogiatzis, senior lecturer in Computer Science at Aston University.

Capable of being set up to view any traffic junction, both real and simulated, the AI starts learning autonomously right away. Other areas can be tweaked as well. For example, the reward system can be manipulated to encourage fast passage for emergency vehicles. Importantly, though, the AI always teaches itself it is never programmed with specific orders.

Ideally, study authors plan on testing the system on real roads this year.

The team presented their findings at the Autonomous Agents and Multi-agent Systems Conference 2022.

See more here:
Traffic lights using artificial intelligence could soon make gridlock a thing of the past - Study Finds

Researchers have created a machine-learning system that efficiently predicts the future trajectories of multiple road users, like drivers, cyclists, and pedestrians, which could enable an autonomous vehicle to more safely navigate city streets. If a robot is going to navigate a vehicle safely through downtown Boston, it must be able to predict what nearby drivers, cyclists, and pedestrians are going to do next. Credit: MIT

A new machine-learning system may someday help driverless cars predict the next moves of nearby drivers, pedestrians, and cyclists in real-time.

Humans may be one of the biggest roadblocks to fully autonomous vehicles operating on city streets.

If a robot is going to navigate a vehicle safely through downtown Boston, it must be able to predict what nearby drivers, pedestrians, and cyclists are going to do next.

Behavior prediction is a tough problem, however, and current artificial intelligence solutions are either too simplistic (they may assume pedestrians always walk in a straight line), too conservative (to avoid pedestrians, the robot just leaves the car in park), or can only forecast the next moves of one agent (roads typically carry many users at once.)

MIT researchers have devised a deceptively simple solution to this complicated challenge. They break a multiagent behavior prediction problem into smaller pieces and tackle each one individually, so a computer can solve this complex task in real-time.

These simulations show how the system the researchers developed can predict the future trajectories (shown using red lines) of the blue vehicles in complex traffic situations involving other cars, bicyclists, and pedestrians. Credit: MIT

Their behavior-prediction framework first guesses the relationships between two road users which car, cyclist, or pedestrian has the right of way, and which agent will yield and uses those relationships to predict future trajectories for multiple agents.

These estimated trajectories were more accurate than those from other machine-learning models, compared to real traffic flow in an enormous dataset compiled by autonomous driving company Waymo. The MIT technique even outperformed Waymos recently published model. And because the researchers broke the problem into simpler pieces, their technique used less memory.

This is a very intuitive idea, but no one has fully explored it before, and it works quite well. The simplicity is definitely a plus. We are comparing our model with other state-of-the-art models in the field, including the one from Waymo, the leading company in this area, and our model achieves top performance on this challenging benchmark. This has a lot of potential for the future, says co-lead author Xin Cyrus Huang, a graduate student in the Department of Aeronautics and Astronautics and a research assistant in the lab of Brian Williams, professor of aeronautics and astronautics and a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL).

Joining Huang and Williams on the paper are three researchers from Tsinghua University in China: co-lead author Qiao Sun, a research assistant; Junru Gu, a graduate student; and senior author Hang Zhao PhD 19, an assistant professor. The research will be presented at the Conference on Computer Vision and Pattern Recognition.

The researchers machine-learning method, called M2I, takes two inputs: past trajectories of the cars, cyclists, and pedestrians interacting in a traffic setting such as a four-way intersection, and a map with street locations, lane configurations, etc.

Using this information, a relation predictor infers which of two agents has the right of way first, classifying one as a passer and one as a yielder. Then a prediction model, known as a marginal predictor, guesses the trajectory for the passing agent, since this agent behaves independently.

A second prediction model, known as a conditional predictor, then guesses what the yielding agent will do based on the actions of the passing agent. The system predicts a number of different trajectories for the yielder and passer, computes the probability of each one individually, and then selects the six joint results with the highest likelihood of occurring.

M2I outputs a prediction of how these agents will move through traffic for the next eight seconds. In one example, their method caused a vehicle to slow down so a pedestrian could cross the street, then speed up when they cleared the intersection. In another example, the vehicle waited until several cars had passed before turning from a side street onto a busy, main road.

While this initial research focuses on interactions between two agents, M2I could infer relationships among many agents and then guess their trajectories by linking multiple marginal and conditional predictors.

The researchers trained the models using the Waymo Open Motion Dataset, which contains millions of real traffic scenes involving vehicles, pedestrians, and cyclists recorded by lidar (light detection and ranging) sensors and cameras mounted on the companys autonomous vehicles. They focused specifically on cases with multiple agents.

To determine accuracy, they compared each methods six prediction samples, weighted by their confidence levels, to the actual trajectories followed by the cars, cyclists, and pedestrians in a scene. Their method was the most accurate. It also outperformed the baseline models on a metric known as overlap rate; if two trajectories overlap, that indicates a collision. M2I had the lowest overlap rate.

Rather than just building a more complex model to solve this problem, we took an approach that is more like how a human thinks when they reason about interactions with others. A human does not reason about all hundreds of combinations of future behaviors. We make decisions quite fast, Huang says.

Another advantage of M2I is that, because it breaks the problem down into smaller pieces, it is easier for a user to understand the models decision-making. In the long run, that could help users put more trust in autonomous vehicles, says Huang.

But the framework cant account for cases where two agents are mutually influencing each other, like when two vehicles each nudge forward at a four-way stop because the drivers arent sure who should be yielding.

They plan to address this limitation in future work. They also want to use their method to simulate realistic interactions between road users, which could be used to verify planning algorithms for self-driving cars or create huge amounts of synthetic driving data to improve model performance.

Predicting future trajectories of multiple, interacting agents is under-explored and extremely challenging for enabling full autonomy in complex scenes. M2I provides a highly promising prediction method with the relation predictor to discriminate agents predicted marginally or conditionally which significantly simplifies the problem, wrote Masayoshi Tomizuka, the Cheryl and John Neerhout, Jr. Distinguished Professor of Mechanical Engineering at University of California at Berkeley and Wei Zhan, an assistant professional researcher, in an email. The prediction model can capture the inherent relation and interactions of the agents to achieve the state-of-the-art performance. The two colleagues were not involved in the research.

Reference: M2I: From Factored Marginal Trajectory Prediction to Interactive Prediction by Qiao Sun, Xin Huang, Junru Gu, Brian C. Williams and Hang Zhao. 28 March 2022, Computer Science > Robotics.arXiv:2202.11884

This research is supported, in part, by the Qualcomm Innovation Fellowship. Toyota Research Institute also provided funds to support this work.

Read the original post:
Predicting Others Behavior on the Road With Artificial Intelligence - SciTechDaily

A doctor cant tell if somebody is Black, Asian, or white, just by looking at their X-rays. But a computer can, according to a surprising new paper by an international team of scientists, including researchers at the Massachusetts Institute of Technology and Harvard Medical School.

The study found that an artificial intelligence program trained to read X-rays and CT scans could predict a persons race with 90 percent accuracy. But the scientists who conducted the study say they have no idea how the computer figures it out.

When my graduate students showed me some of the results that were in this paper, I actually thought it must be a mistake, said Marzyeh Ghassemi, an MIT assistant professor of electrical engineering and computer science, and coauthor of the paper, which was published Wednesday in the medical journal The Lancet Digital Health. I honestly thought my students were crazy when they told me.

At a time when AI software is increasingly used to help doctors make diagnostic decisions, the research raises the unsettling prospect that AI-based diagnostic systems could unintentionally generate racially biased results. For example, an AI (with access to X-rays) could automatically recommend a particular course of treatment for all Black patients, whether or not its best for a specific person. Meanwhile, the patients human physician wouldnt know that the AI based its diagnosis on racial data.

The research effort was born when the scientists noticed that an AI program for examining chest X-rays was more likely to miss signs of illness in Black patients. We asked ourselves, how can that be if computers cannot tell the race of a person? said Leo Anthony Celi, another coauthor and an associate professor at Harvard Medical School.

The research team, which included scientists from the US, Canada, Australia, and Taiwan, first trained an AI system using standard datasets of X-rays and CT scans, where each image was labeled with the persons race. The images came from different parts of the body, including the chest, hand, and spine. The diagnostic images examined by the computer contained no obvious markers of race, like skin color or hair texture.

Once the software had been shown large numbers of race-labeled images, it was then shown different sets of unlabeled images. The program was able to identify the race of people in the images with remarkable accuracy, often well above 90 percent. Even when images from people of the same size or age or sex were analyzed, the AI accurately distinguished between Black and white patients.

But how? Ghassemi and her colleagues remain baffled, but she suspects it has something to do with melanin, the pigment that determines skin color. Perhaps X-rays and CT scanners detect the higher melanin content of darker skin, and embed this information in the digital image in some fashion that human users have never noticed before. Itll take a lot more research to be sure.

Could the test results amount to proof of innate differences between people of different races? Alan Goodman, a professor of biological anthropology at Hampshire College and coauthor of the book Racism Not Race, doesnt think so. Goodman expressed skepticism about the papers conclusions and said he doubted other researchers will be able to reproduce the results. But even if they do, he thinks its all about geography, not race.

Goodman said geneticists have found no evidence of substantial racial differences in the human genome. But they do find major differences between people based on where their ancestors lived.

Instead of using race, if they looked at somebodys geographic coordinates, would the machine do just as well? asked Goodman. My sense is the machine would do just as well.

In other words, an AI might be able to determine from an X-ray that one persons ancestors were from northern Europe, anothers from central Africa, and a third persons from Japan. You call this race. I call this geographical variation, said Goodman. (Even so, he admitted its unclear how the AI could detect this geographical variation merely from an X-ray.)

In any case, Celi said doctors should be reluctant to use AI diagnostic tools that might automatically generate biased results.

We need to take a pause, he said. We cannot rush bringing the algorithms to hospitals and clinics until were sure theyre not making racist decisions or sexist decisions.

Hiawatha Bray can be reached at hiawatha.bray@globe.com. Follow him on Twitter @GlobeTechLab.

Here is the original post:
MIT, Harvard scientists find AI can recognize race from X-rays and nobody knows how - The Boston Globe

Technology for MSME: The shift towards more efficient technology solutions from the good old websites and emails, in the name of digital adoption, is apparent among MSMEs that have shied away from evolving technologies for a long time. The shift has largely been visible because of better affordability due to growing on-demand, or pay as you go or what is called the software-as-a-service (SaaS) ecosystem in India liberating small businesses from the cost conundrum to some extent.

As India observes the National Technology Day on Wednesday to commemorate its entry into the elite club of countries having nuclear weapons with the Pokhran nuclear tests in 1998, it is also the day to remember the countrys achievement in science and innovation. While a large number of MSMEs are yet to fully benefit from the technology revolution, some of them have certainly been warming up to the new age solutions such as artificial intelligence (AI) and using it also for better growth.

The implementation of AI is across multiple use cases. For instance, Delhi-based long-haul logistics services provider JCCI Logistics has deployed AI and internet of things (IoT) solutions to manage its fleet of around 150 trucks. The company, launched in 2004, uses on-demand fleet management software for GPS tracking of vehicles, fuel management, driver analytics, and route planning.

Vehicles need to run as much as possible and thats what matters. Before deploying this solution in 2020, our monthly cumulative running was around 8,000 to 10,000 kilometres. It has increased by around 20 per cent now. The jump, I think, is primarily because of the on-board diagnostics (OBD) device that you can fit in a vehicle to get data related to fuel consumption, drivers driving behaviour, whether there is unnecessary hard acceleration or not, etc., Sachin Jain, Founder, JCCI Logistics told Financial Express Online.

OBD is essentially a machine learning (ML) and internet of things (IoT) based device that gets signals from different sensors in a vehicle and conveys them to the users dashboard with the help of the software.

JCCI Logistics have been among post-Covid adopters of deep technology solutions as the pandemic perhaps necessitated the use of software and digital for sustenance.

Covid might have caused a faster switch to some AI/ML applications since the labor force was locked up. AI/ML provides a significant opportunity for reduction in input costs, particularly those of human capital. The advent of edge AI/ML will further hasten adoption, particularly as it gets married to IoT on small devices and sensors that are available at scale and used routinely by businesses of all sizes, Utkarsh Sinha, Managing Director at advisory firm Bexley Advisors told Financial Express Online.

Among the top sectors where the use of AI accelerated during the pandemic was restaurant as the pandemic precipitated the eateries into looking at ways that could help them optimize their processes right from sales to inventory management and more.

Kabir Suri who runs Azure Hospitality, which owns restaurant chains like Mamagoto, Dhaba, Speedy Chow, etc., has been using AI in the companys operations for the past five years while Covid only reinforced his commitment to AI for efficiency and growth. We have had a direct saving of 30 per cent in past five years along with getting customer insights due to AI that has led to an uptake in revenue as well. Five years back we had around 10 outlets and now have 60 across India, Suri told Financial Express Online.

Subscribe to Financial Express SME newsletter now: Your weekly dose of news, views, and updates from the world of micro, small, and medium enterprises

The company has an in-house AI solution that shows live sales, total transactions, menus, items sold, total consumption per restaurant, etc. The solution captures data from every restaurant throughout the day on a real time basis and consolidates it to show up for analysis on its dashboard. This becomes important for restaurants with chains to understand the consumer-behaviour pattern, the impact of different occasions on business like festivals such as Navratras in North particularly, Christmas in Goa, and some other festivals in South, said Suri.

Moreover, the AI solution at Azure Hospitality helps Suri control the HR module as well. You can look at your salary component, leaves, attendance, holidays, payslips, etc., through a single system every day whenever you want. Basically, AI helps you make better decisions as you grow bigger by minimizing the impact of any uncertainty, Suri added.

Another sector that depends heavily on technology and AI particularly is tourism for purposes ranging from travel booking via chatbots, flight forecasting in terms of the current best price and future prices, recommendations for hotel and cab booking based on travel-related searches, etc.

There is AI at every stage in tourism and aviation, Subhas Goyal, Founder and Chairman at B2B travel company STIC Travel told Financial Express Online. The company is the exclusive General Sales Agent (GSA) a sales representative of a company in a specific region or country for 11 international airlines in India including United Airlines, Air China, Croatia Airlines etc.

STIC has been using for the past five years AI-based Microsoft Dynamics CRM to manage customer relationships, track sales leads, marketing, etc., and streamline administrative processes in sales and marketing. The company is now also implementing a chatbot assistant to answer customer queries on its platform. Goyal noted the standard queries around bookings, holiday searches can be answered by the AI bot while for further details and feedback, there would be manual intervention.

Post-Covid, more MSMEs had started to use primary technology tools at least such as social media, online service aggregators, company websites etc. According to a Crisil survey of around 540 micro and small units released in April this year, over 65 per cent respondents adopted or upgraded their use of online aggregators, social media platforms, and company websites. Among sectors, manufacturing reported higher adoption with 71 per cent respondents adopting or upgrading their use of digital platforms in comparison to 66 per cent respondents in the services sector.

Good technology is invisible. AI/ML will soon form a fundamental layer in all operations and interactions for small businesses. As technology offerings scale, it will soon be easier to get good AI to do certain tasks than to get a human to do it. The impact of this on labor force utilization will be significant, added Sinha.

Here is the original post:
National Technology Day: How artificial intelligence is helping MSMEs to optimize processes, accelerate growth - The Financial Express