Safer Living Through AI and IoT
The world is increasingly afflicted by natural disasters. Almost every day we turn on the news to see fires, floods, hurricanes, tsunamis and other storms striking yet another major population center. By now, many of us—or our families and friends—have been personally affected. And while in most cases we can’t yet prevent these occurrences, we can begin to better prepare for them and mitigate damage.
Disaster management is a very real area of research that predates much of today’s technology, and is one that is eager to embrace its potential. Experts in this area proffer three key pieces of advice: take measures to mitigate potential damage, implement means for immediate victim assistance and plan for rapid recovery. While these pieces of advice were probably originally conceived at a time when the main actors in disaster management would be people, technology can and is now helping with all three.
In some ways, huge trends, such as AI, the IoT and Big Data, have the intensity of natural phenomena, but they have the potential to be forces for good. We can now use technology to spread alerts faster than ever before, ensuring people living in areas of risk can be better prepared to take evasive action should the need arise. Smart sensors can now supply the raw data needed to detect potential threats sooner, and high-speed networks can deliver sensor data to server farms where AI can crunch the numbers to find patterns that match threats.
But there is much work to be done. The global financial impact of natural disasters has been estimated at more than $300 billion a year and climbing, with some estimates much higher when taking downstream impacts into account. Unfortunately, according to the United Nations 2019 Global Assessment Report on Disaster Risk Reduction (GAR2019) , today’s international development financing system allocates approximately 20 times the funding to emergency response, reconstruction, relief and rehabilitation activities compared to that allocated for disaster prevention and preparedness.
So how will IoT technologies help with prevention and preparedness? The IoT is pervasive, and its technology is becoming less expensive, which makes endpoints like smart sensors more cost-effective and relatively easy to deploy (Figure 1). In terms of early warning systems, we can expect more raw data to be generated in areas prone to natural disaster through various sensors to measure earth tremors, monitor sea levels, measure carbon monoxide/dioxide levels, monitor temperature and moisture levels and more. Changes in such elements can forewarn us of imminent danger.
The data generated by each of these sensors is the key. According to the GAR2019 report, today data collection is “…often fragmented, non-universal, incommensurable and biased, and the disconnect among ‘knowing’ something, making it ‘available and accessible’ and ‘applying’ what is known, often remains.” We see this same macro issue at work in local environments where IoT sensors are being used by municipalities and companies to gather data from various systems to create better living for citizens and employees.
Growing populations across the world are increasingly migrating to cities—many of which are rapidly turning into megacities with populations greater than 10 million people. In this environment, access to data becomes fundamental to safer living. In today’s cities, IoT technology can approximate how long it will take us to drive across the city, warn us of road accidents, map our route, and help us find a parking space.
But as populations in urban areas increase, the disproportional gulf between cause and effect will become more apparent. For example, road traffic incidents may be attributable to a build-up of traffic in another part of the city, or the lack of adequate lighting on a particular street. More densely inhabited areas may generate greater potential for incidents, because the margins for error will be eroded. If one person takes a different route home it makes little or no difference to congestion; if a hundred people do it, roads can become gridlocked.
This is where the integration of disparate systems and use of AI will make all the difference. In the future, real-time data, forming seemingly incoherent patterns, will be easily analyzed by AI technologies to make traffic flow better, or reduce the potential of hazards for pedestrians and cyclists. And it will happen behind the scenes, without us having to make a conscious effort to change our natural behavior. Right now, the systems to make this work aren’t seamless. These systems, even those that are connected to the internet, often exist in isolated silos.
SMART BUILDING EXAMPLE
Take a smart building as an example. Within a building, the access control systems, HVAC systems, lighting systems, elevators and other systems may all be “smart” in that they automatically turn on when needed, turn off when they aren’t needed, can be monitored and adjusted remotely, but they are generally disconnected from each other. It won’t be much use during an emergency if a building’s emergency lights turn on, but the doors remain locked.
The reason for the disconnect is largely a legacy issue: there are so many different, un-interoperable protocols, devices and services used in existing building management systems and other industrial control systems, that integration has become a real issue. Where these systems are able to connect and work together today, it often takes vast sums of money to fund the integration effort. What we need are simple, cost-effective ways to bridge legacy systems to new IoT systems to let us make use of the valuable data that the systems generate.
The SmartServer IoT from Adesto is designed to address this issue. It makes it easier to access the wealth of data an industrial control system may hold, to enable new solutions that could make a real difference to peoples’ lives. With SmartServer IoT, companies can easily connect their disparate, non-interoperable systems, devices, and services together and also connect to cloud platforms to make use of AI and predictive analytics—which can be used to understand trends and mitigate risks (Figure 2).
Natural disasters are potentially predictable, and manufactured incidents are often avoidable. Both rely on being able to observe, analyze and react to the world around us. And while global natural-disaster risk mitigation will require mega political, socioeconomic and cultural discussions and change, the AI and IoT technologies that can enable this change are increasingly available.
Today’s technology means that we are now more equipped—through data—to defend ourselves, our homes and our possessions from harm. In the future, by bringing disparate systems together and making existing solutions more extensible, we can build even smarter and safer communities.
For detailed article references and additional resources go to: www.circuitcellar.com/article-materials
Reference  as marked in the article can be found there.
Adesto Technologies | www.adestotech.com
PUBLISHED IN CIRCUIT CELLAR MAGAZINE • JANUARY 2020 #354 – Get a PDF of the issueSponsor this Article
Jen Bernier-Santarini is VP of Corporate Communications at Adesto, a provider of application-specific semiconductors and systems for IoT. Before joining Adesto in 2019, Jen led technology communications for IP provider Imagination Technologies. With more than 25 years working in semiconductors and related technologies, her expertise includes electronic design automation (EDA) tools, connectivity technologies, processors and IP, flash memory and other off-the-shelf chips.