The future when people will come home after work and ask their TV to
turn on and the washing machine to wash clothes in the economy mode doesn’t
seem so distant.
We can already talk to virtual assistants like Google Home, Siri or
Alexa to search for a movie or order a new scarf with delivery at door. Why not
doing the same thing with everything else?
In fact, this is what everyone now calls the Internet of Things which
is according to Wikipedia basically the network of physical devices, vehicles,
home appliances, and other items embedded with electronics, software, sensors,
actuators, and connectivity which enables these things to connect, collect and
exchange data. For any IoT service to be worth buying, such actions must
demonstrate true value and yield benefit to the user. Of course, they vary from
adequate physical actions (e.g. deploying a taxi to the site) to simply
informing users (e.g. sending a message to inform a user that they have run out
of milk).
It is here at the data analysis step that the true value of any IoT
application is determined, and this is where Artificial Intelligence provides a
crucial role by making sense of data streamed from devices. AI serves to detect
patterns in this data from which it can learn to adjust the behavior of IoT
service.
Probably the best example of AI and IoT successfully working together
is self-driving cars by Tesla Motors. Cars act as “things” and use the power of
Artificial Intelligence to predict the behavior of cars and pedestrians in
various circumstances. Moreover, all Tesla cars operate as a network. When one
car learns something, they all learn it.
Automated vacuum cleaners are a good example of artificial intelligence
“embodied” in a robot. For example, iRobot by Roomba controlled through an app
can map and “remember” a home layout, adapt to different surfaces or new items,
clean a room with the most efficient movement pattern, and dock itself to
recharge its batteries.
Another good example of AI and IoT combined together is a smart
thermostat solution by Nest Labs. Nest’s smartphone integration allows to check
and control temperature from anywhere. The device analyzes temperature
preferences and work schedule of its users and adapts temperature accordingly.
Applications, where IoT works together with AI, are only growing,
creating new markets and opportunities and they are highly unlikely to lose
ground in the nearest future.
The Internet of Things is getting smarter. Companies are incorporating
artificial intelligence—in particular, machine learning—into their IoT
applications. The key: finding insights in data.
With a wave of investment, a raft of new products, and a rising tide of
enterprise deployments, artificial intelligence is making a splash in the
Internet of Things (IoT). Companies crafting an IoT strategy, evaluating a
potential new IoT project, or seeking to get more value from an existing IoT
deployment may want to explore a role for AI.
Artificial intelligence plays a growing role in IoT applications and
deployments. Both investments and acquisitions in startups that merge AI and
IoT have climbed over the past two years. Major vendors of IoT platform
software now offer integrated AI capabilities such as machine learning-based
analytics.
The value of AI in this context is its ability to quickly wring
insights from data. Machine learning, an AI technology, brings the ability to
automatically identify patterns and detect anomalies in the data that smart
sensors and devices generate—information such as temperature, pressure,
humidity, air quality, vibration, and sound. Compared to traditional business
intelligence tools—which usually monitor for numeric thresholds to be
crossed—machine learning approaches can make operational predictions up to 20
times earlier and with greater accuracy.
Other AI technologies such as speech recognition and computer vision
can help extract insight from data that used to require human review.
In its essence, the technology of IoT is about devices with built-in
sensors, which provide data to one or more central locations through internet
connectivity. That data is then analyzed and corresponding actions are
initiated. AI applications for IoT enable companies to avoid unplanned
downtime, increase operating efficiency, spawn new products and services, and
enhance risk management.
AVOIDING COSTLY UNPLANNED DOWNTIME
In a number of sectors—industrial manufacturing or offshore oil and
gas, to name two—unplanned downtime resulting from equipment breakdown can cost
big money.
Predictive maintenance—using analytics to predict equipment failure
ahead of time in order to schedule orderly maintenance procedures—can mitigate
the damaging economics of unplanned downtime. Machine learning makes it
possible to identify patterns in the constant streams of data from today’s
machinery to predict equipment failure. In manufacturing, Deloitte finds
predictive maintenance can reduce the time required to plan maintenance by
20–50 percent, increase equipment uptime and availability by 10–20 percent, and
reduce overall maintenance costs by 5–10 percent.
INCREASING OPERATIONAL EFFICIENCY
AI-powered IoT can also help improve operational efficiency. Just as
machine learning can predict equipment failure, it can predict operating
conditions and identify parameters to be adjusted on the fly to maintain ideal
outcomes, by crunching constant streams of data to detect patterns invisible to
the human eye and not apparent on simple gauges.
Machine learning often finds counterintuitive insights: A shipping
fleet operator’s machine learning tools determined that cleaning their ships’
hulls more often—an expensive, downtime-causing process—actually increased the
fleet’s overall profitability. The math went against shipping industry
instincts: Hulls kept smooth through frequent cleaning improve fuel efficiency
enough to vastly outweigh the increased cleaning costs.
ENABLING NEW AND IMPROVED PRODUCTS AND SERVICES
Enhancing IoT with AI can also directly create new products and
services. Natural language processing (NLP) is getting better and better at
letting people speak with machines, rather than requiring a human operator.
AI-controlled drones and robots—which can go where humans can’t—bring all-new
opportunities for monitoring and inspection that simply didn’t exist before.
Fleet management for commercial vehicles is being reinvented through AI,
which can monitor every measurable data point in a fleet of planes, trains,
trucks or automobiles to find more efficient routing and scheduling, and reduce
unplanned downtime. Cloudera claims its fleet management AI has cut downtime
for fleet vehicles monitored by Navistar devices up to 40 percent.
ENHANCING RISK MANAGEMENT
A number of applications pairing IoT with AI are helping organizations
better understand and predict a variety of risks as well as automate for rapid
response, enabling them to better manage worker safety, financial loss, and
cyber threats.
Applications already in use include detecting fraudulent behavior at
bank ATMs, predicting auto driver insurance premiums based on their driving
patterns, identifying potentially hazardous stress conditions for factory
workers, and monitoring law enforcement surveillance data to identify likely
crime scenes ahead of time.
IMPLICATIONS FOR ENTERPRISES
For enterprises across industries, AI is a natural complement to IoT
deployments, enabling better offerings and operations to give a competitive
edge in business performance.
Machine learning for predictive capabilities is now integrated with
most major general-purpose and industrial IoT platforms, such as Microsoft
Azure IoT, IBM Watson IoT, Amazon AWS IoT, many more
A growing number of turnkey, bundled, or vertical IoT solutions take
advantage of AI technologies, especially machine learning. It is often possible
to use AI technology to wring more value from IoT deployments that were not
designed with the use of AI in mind. IoT deployments generate huge, constant
streams of data, which machine learning excels at examining to identify
patterns that lead to greater value.
THE FUTURE OF IoT IS AI
It may soon become rare to find an IoT implementation that does not
make some use of AI. The International Data Corp. predicts that by 2019, AI
will support “all effective” IoT efforts and without AI, data from the
deployments will have “limited value.” If your company has plans for
implementing IoT-based solutions, those plans should probably include AI as
well.
The Internet of Things (IoT) is a term that has been introduced in
recent years to define objects that are able to connect and transfer data via
the Internet. ‘Thing’ refers to a device which is connected to the internet and
transfers the device information to other devices. The cloud-based IoT is used
to connect a wide range of things such as vehicles, mobile devices, sensors,
industrial equipment’s and manufacturing machines to develop a various smart
systems it includes smart city and smart home, smart grid, smart industry,
smart vehicle, smart health and smart environmental monitoring. In the IoT,
cloud computing environment has made the task of handling the large volume of
data generated by connecting devices easy and provides the IoT devices with
resources on-demand.
An increasing number of physical objects are being connected to the
Internet at an unprecedented rate realizing the idea of the Internet of Things
(IoT). A recent report states that “IoT smart objects are expected to reach 212
billion entities deployed globally by the end of 2020”. Similarly, while the
number of connected devices already exceeds the number of humans on the planet
by over 2 times, for most enterprises, simply connecting their systems and
devices remains the first priority. A recent report state that, “The overall
Internet of Things market is projected to be worth more than one billion U.S.
dollars annually from 2017 onwards”. As a result, data production at this stage
will be 44 times greater than that in 2009, indicating a rapid increase in the
volume, velocity and variety of data.
Hence, IoT based smart systems generate a large volume of data often
called big data that cannot be processed by traditional data processing
algorithms and applications. Here will therefore, by difficulty in storing,
processing and visualizing this huge data generated from IoT based system.
However, there is highly useful information and so many potential values hidden
in the huge volume of IoT based sensor data. IoT based sensor data has gained
much attention from researchers in healthcare, bioinformatics, information
sciences, policy and decision makers in governments and enterprises. Nowadays,
Artificial intelligence methods play a significant role in various environments
including business monitoring, healthcare applications, production development,
research and development, share market prediction, business process, industrial
applications, social network analysis, weather analysis and environmental
monitoring.
The IoT and Artificial Intelligence (AI) will play a vital role in
numerous ways in the future. There are multiple forces which are driving the
growing need for both technologies and more and more industries, governments,
engineers, scientists and technologists have started to implement it in
manifold circumstances. The potential opportunities and benefits of both AI and
IoT can be practiced when they are combined, both at the devices end as well as
at server. For example, AI combined with Machine learning can study from the
data to analyze and predict the future actions in advance, such as order
replacements in marketing and failure of equipment in an industry just in time.
Moreover, AI can be used with machine learning in smart-homes to make a truly
grand smart home experience. Similarly, AI methods with IoT can be used to
analyze the human behavior via Bluetooth signals, motion sensors, or
facial-recognition technology and to make the corresponding changes in lighting
and room temperatures. This special issue aims to gather recent research works
in emerging artificial intelligence methods for processing and storing the data
generated from cloud-based Internet of Things.
