Summary artificial intelligence in practice- part-5

Some Impressionistic Take away from the Book of
Bernard Marr & Matt Ward
Artificial Intelligence in Practice
( Part – 5)
( How 50 Successful companies used AI & Machine Learning to Solve problems)
Ramki
ramaddster@gmail.com

The Summary of this book is made in 4 parts
due to large coverage of the book .
This is Part – 5
( Read this after Part-1 , 2 ,3, 4)

Part – 5
Manufacturing, Automotive, Aerospace &
Industry 4.0 Companies

BMW
Using AI to Build & drive the Cars of tomorrow

 BMW creates some of the most high-tech cars we have yet seen. The
German giant builds 2.5 million vehicles every year and sells them all over
the world.
 But technology is not just limited to the cars it builds, its business model is
built on Big Data which drives everything it does across design, engineering,
production, sales and customer support.
 Thanks to AI, data-driven predictive analytics and other cutting edge
technologies BMW is able to build the cars of today while at the same time
envisaging and bringing to reality the cars of tomorrow.
 BMW is clearly confident in its belief that cars of the near future will pilot
themselves, rather than relying on human drivers. All the big auto
manufacturers are staking their claim in a driverless future, but BMW has
done so with more confidence than most. It has stated that its aim is for its
vehicles to achieve full “level 5” autonomy by 2021.
 Level 5 Autonomy tops the scale defined by the US Department of Transport
as it made preliminary investigations into how legislation for autonomous
vehicles would work. It indicates that the car will be capable of driving with no
human input or supervision and operate at least as effective as a human
driver in any conditions and on any road.
BMW

 More details of how this would be achieved began to emerge earlier
this year when BMW announced a partnership with Intel, which itself
had recently acquired Mobileye, a leader in computer vision
technology.
 Computer vision is based on the idea of teaching machines to “see” in
the same way humans do, using cameras instead of biological eyes
and to interpret the information in a similar way to our brains.
 It is an advanced form of image recognition – which can be seen in
action in Google Image Search as well as many other ML applications,
where machines have been taught to sort and classify images,
becoming more adept as they are exposed to more and more data.
 Of course, rather than sorting harmless images of cats and dogs on
the internet, the computer vision used in autonomous driving will have
to be capable of reading all of the input data from the cars’ cameras
and sensors and analyzing it in real-time - quickly enough to take
emergency action at 100 kph.
BMW & Mobile eye

 One group of people for whom the concept of not having to drive
themselves around is nothing new is Rolls Royce owners.
 Unsurprisingly, this brand (also owned by BMW) styles its self-driving
software as a virtual chauffeur. At the controls of the concept Model
103EX is Eleanor, named after the actress and model believed to have
been the inspiration for the cars’ famous Spirit of Ecstasy hood
ornaments.
 The body is made from one seamless, molded piece of metal, in the form
of sculpture and the car generates a personal red carpet using LEDs for
when its occupant steps out.
 Sure, this might not be the sort of car most of us will own in 10 years, but
the design of the 103EX gives us some clues about how the super-rich
will be served by AI and autonomy.
 For other applications which are likely to enjoy a wider exposure among
its customers, BMW has partnered with a number of other leaders in the
AI field, including IBM. Its Watson cognitive computing platform was used
in prototype i8 hybrid vehicles to learn about how drivers and their cars’
systems can interact more comfortably and naturally.
Automated Chauffeurs

 Through its partnership with location data service provider Here (which BMW co-
owns along with Volkswagen and Daimler after acquiring it from Nokia last year),
data is already been collected which could help educate the first wave of
consumer-ready self-driving cars
 Data including video from onboard cameras, machine data such as braking force,
wiper and headlight use, and GPS information, is already being scooped up by
vehicles and fed into Here to help with mapping and route planning in today’s
cars.
 Tomorrow it could also be used to train them how they should react and behave
when they become fully autonomous.
 Another innovation which is in use today but could form the foundation of more
complex services in the future is shown through its partnership with Parkmobile,
which offers mobile payment for parking services across the US.
 A recently announced deal will see the systems installed as standard across a
range of BMW models. It will allow drivers to find and pay for parking spaces at
their destinations before they set off. As well as convenience, this has the
advantage of cutting carbon waste as the average driver spends 20 minutes
searching for his or her parking space on each trip to somewhere unfamiliar. In the
future when vehicles are autonomous it is planned that these systems will allow
machine-to-machine payments to take place transparently to the driver.
Location Analytics

 Self driving cars are seen by every major automobile
manufacturer as the future of personal transportation.
 The autonomous cars of the future will be safer & more
efficient thanks to their use of AI to anticipate & react to
unexpected circumstances on the road.
 Traditional car companies are partnering with tech companies
to bring in the expertise needed when it comes to integrating
cutting-edge cognitive software with large-scale production.
 We are likely to see far greater integration of AI manually
driven cars in the form of virtual assistant AIs, changing the
way we operate & interact with vehicles.
Results, Key Challenges, Learning points & Takeaways

GE-
Using AI to Build The Internet of Energy

 Founded by Thomas Edison & today operates globally –
Power, manufacturing, healthcare, aviation, oil & gas ,
financial sector.
 Challenges- Data- driven, Digital revolution & need for a
switch to more sustainable sources of energy.
 The global energy industry is facing disruption as it transitions
from fossils to renewables (and occasionally back again). Its
challenges include balancing growing demand in developing
nations with the need for sustainability, and predicting the
effect of extreme weather conditions on supply and demand.
 GE has invested over a billion dollar in 5 years transforming
itself from an industrial company to a software & analytics
company – Smart and self learning machines.
GE

 GE Power – whose turbines and generators supply 30 percent of the
world’s electricity – has been working on applying Big Data, ML &
IoT technology to build an “internet of energy” to replace the traditional,
linear, one-way model of energy delivery.
 Ganesh Bell, first and current chief data officer at GE Power, -“If you think
about it, the electricity industry is still following a one-hundred-year-old
model which our founder, Edison, helped to proliferate. It’s the generation
of electrons in one source which are then transmitted in a one-way linear
model … That whole infrastructure is now being tested and pushed every
day because of the challenges we’re talking about.”
 The answer to these challenges, Bell believes, lies in taking advantage of
the networked, grid-based generation and delivery infrastructures, while
augmenting it with the flow of data. “We think of a world where every
electron will have a data bit associated with it, and we associate and track
that data and optimize it, and suddenly, from a linear model, we have
moved to a networked model,” says Bell. It certainly makes sense in a
world where everything is increasingly becoming networked and
connected – from the devices in our homes to transport networks.
How GE uses Technology in Practice

 Functions enabled by advanced analytics and machine learning, such as
predictive maintenance and power optimization, can then be applied to
critical infrastructure machinery.
 As Bell says, “We have seen results like reducing unplanned downtime by 5
percent, reducing false positives by 75 percent, reducing operations and
maintenance costs by 25 percent – and these start adding up to meaningful
value.”
 As well as asset performance management, GE categorizes its data-powered
applications into two other groups – one is operations optimization, which
focuses on insights that can be applied across a whole plant, or enterprise.
 And the other is business optimization – applications designed to improve the
profitability of customers, “So they can use weather data, energy market
pricing data, lots of internal and external data to make sure they are
capturing every opportunity for optimizing their business and being more
profitable.”
 Put together, these three categories of application make up the foundations
of GE Power’s vision for the “digital power plant” – the first step towards
making the internet of energy a possibility.
How GE uses Technology in Practice

 The power industry demand increase is over 50% in next 20 years-
cutting carbon footprint by 50%.
 Data feeds directly into applications such as GE Power’s own asset
performance management software, which enables equipment to be
monitored even if it’s from a third-party manufacturer, meaning it
covers every machine in a power plant, whether or not it’s
manufactured by GE.
 Advanced analytics powered by AI has the potential to help with this
mission.
 Predicting peaks & troughs in the demand for energy within a
geographic region means increased efficiency & less waste.
 Interpretation of data and drawing insights is the real value .
 Datafication is, along with decentralization (the move towards
generating power close to where it will be used) and decarburization
(the move away from fossil fuels), one of the three “Ds” disrupting the
energy industry. And with that data comes great value.

John Deere
Using AI to reduce pesticide pollution in Agriculture

 Founded by a small town blacksmith as a toolmaker & over 150 years
later has become global leading manufacturers & suppliers of
agricultural & industrial machinery.
 Technological innovator- GPS technology in late 1990s.
 Has transformed into a technology company – selling data as a
service to allow farmers to make better decisions for their operations.
 Offering autonomously driving tractors, intelligent sensors & software
& agricultural drones.
 In just 30 years’ time, it is forecasted that the human population of our
planet will be close to 10 billion.
 Producing enough food to feed these hungry mouths will be a
challenge, and demographic trends such as urbanization, particularly
in developing countries, will only add to that.
 This leads to efficient use of the land – increase use of fertilizers.
Environmental risks, direct hazards – concern on human health.
John Deere

 Developed ML technology designed to ensure that where herbicides &
pesticides are used, they are used as sparingly as possible.
 Vastly has cut down the waste, reducing energy usage & environmental
impact of pesticide production.
 Less pollution of local rivers & waterways and optimization of food
production.
 Company uses Blue River technology- acquired in 2017.
 Harness computer vision techniques to sense where crops are threatened
by pests, and control robotic equipment capable of firing accurate blasts
of pesticide chemicals at the affected crops while leaving others
untouched.
 Vast database of crop photographs- used computer vision algorithms to
determined which phots showed crops that were affected.
 Compares with database and takes action.
 Cutting edge AI technology and also Farmsight – data driven decisions
about where and when the crops to be planted.
How AI used in Practice

 Precision agricultural system will reduce the quantity of
pesticides sprayed by 90%.
 Means higher crop yields for farmers and reduce manpower .
 Advanced AI will provide solutions for higher food production
for the world.
 Precision agriculture means reduction in amount of harmful
pesticides – reduction in pollution.
 Automation along with sensing and decision making
technology will break new grounds.
 Teaching automated systems to recognize the difference
between afflicted and healthy crops – with vast amount of
photographic data.

Kone
Using AI to move millions of people everyday

 KONE’s mission is to improve the flow of urban life.
 The Finnish-headquartered elevator and escalator engineering and
maintenance company is responsible for 1.1 million elevators
worldwide.
 As well as offices and apartments, it runs people-moving machinery at
airports, stadiums and exhibition venues. At Heathrow airport in
London alone, it is responsible for moving 191,000 people daily, using
1,035 escalators, elevators and auto walks.
 This is a huge responsibility since breakdowns or faulty equipment can
cause delays affecting thousands of people.
 In recent years, KONE has intensively ramped up efforts to capture as
much data as possible from this machinery and put it into the cloud.
 From here their analysts, backed with sophisticated AI and machine
learning courtesy of IBM’s Watson Platform, can ensure that
everything is operating efficiently, and technicians can diagnose and
react to problems as soon they are predicted to occur, rather than
having to wait and act after something goes wrong.
Kone

 CEO Henrick Ehrnrooth says “We are connecting elevators and
escalators to the cloud – over the next few years we are planning to
collect more than a million of them.
 Of course with Internet of Things and cloud, that means we’re collecting a
lot of data, and this enables us to provide significant value for our
customers.“
 “The key technology – sensoring – it’s out there.
 A lot of people are doing it. We’re taking it further by focusing on how we
can improve the business of our customers.
 When you’re managing a building it’s important to have a full
understanding of what’s going on, all the time – what is happening? How
is the equipment performing? How are people moving in the building?
 KONE’s system – which can interface directly with the company’s own
escalators and elevators to access machine data directly, or use sensors
to collect data from machinery installed by other manufacturers – gives
real-time readouts on everything from start and stop times, to
acceleration, temperature, noise levels and vibrations running through
cables.
How AI used in Kone

 The site takes machine data as elevators communicate through the cloud
and translates it into human speech – it’s an interesting glimpse into what
the idea that machines are now capable of talking to each other really
means!
 KONE announced that it will package this data and provide it to customers
– lift and elevator operators – with the name 24/7 Connected Services.
 A limited amount of Edge computing is also carried out – with some
decisions about what data is or isn’t useful made at the point where it is
collected from the machinery itself. This cuts down on overall data volume
by eliminating useless “noise” at the source.
 With this basic functionality in place, KONE has just completed the first
step of the journey that it is hoping to take with machine learning and
predictive analytics.
 “Now that we have the Watson IoT platform in place, this is just the start”,
Ehrnrooth says. “Now we have the elevators connected to the cloud, and
we are connected to our customers, we can very easily add different
services and parts to the package.”
How AI used in Kone

 The plan is that the Watson system will soon be able to act as a real-time
adviser to technicians, equipped with hand-held terminals and able to get
instant insights to any problems they come up against during their day-to-
day routines.
 “They will be told ‘there is X% likeliness that this is the problem, and this is
how you should fix it’, and that assistance will come from what the system
has learned while it’s been running, and the technical documentation
that’s been fed into it.”
 Other ideas include deploying Watson into call centers, where it will
analyze customer telephone conversations, offering customer service staff
real-time help in the same way it does for technicians.
 Looking perhaps a little further into the future, the idea of “personalizing”
elevator rides is on KONE’s agenda. Just as the wealth of newly available
and capturable data on individuals is enabling personalized shopping and
healthcare, so could it add value during the time we spend going up and
down between floors.

Daimler AG
From Luxury Personal Cars to Passenger Drones

 German parent company of Mercedes-Benz- Luxury cars &
consumer vehicles.
 Renowned precision engineered automobile brand.
 Investing heavily on automation & 4th industrial revolution
technology- design to productions & to vehicles.
 Using AI to create efficiencies in vehicle production,
transportations and passenger transport.
 Design and Manufacturing is labour intensive & costly process
 Equipment breakdown & human error will lead to wasted
resources , costly delays & injury.
 Car ownership to ride sharing, public transport means declining
customer base.
 Changing urban landscapes – less friendly owners & operators
personal vehicles.
Daimler AG

 Daimler has unveiled the Future truck 2025 – self driving heavy
good vehicle.
 Has a cabin to transport a crew and can navigate fully
autonomously – improved road safety & lower fuel cost.
 Benz is investing in AI car – Known as MBUX to free up drivers
from repetitive or distracting tasks while on road.
 Active climate control via indirect commands such as when
driver says “ It is hot”.
Daimler AG

 Daimler appears to be fully committed to the 4th Industrial
Revolution when you look at the digitalization of its factories, sales
and customer experience efforts—all made possible through the
technology of big data and machine learning.
 Digitalization of factories allows Daimler to have real-time
knowledge of supplies, efficiently respond and adjust to the
customization required for each vehicle as demanded by today’s
consumer and helps Daimler be flexible, efficient, quick and smart
in its production of cars.
 Every element of Daimler’s production is impacted by digitalization
from digital design and prototyping to networked production. The
company is showing how humans and robots can co-exist and
enhance production.
 Because production is connected, the customer gets to joyfully
anticipate their new car and follow it from contract to creation along
the production route with a new tool called Joyful Anticipation.
Daimler AG

 Most of Daimler’s AI projects are in Pilot or prototype stage .
 Company is focused on intelligent, self-learning technology .
 Company partnered with Nvidia to design deep learning based
systems 4 years ago.
 Production environments data is gathered from cameras &
sensors fitted with machinery as well as data from computerized
stock control systems, machine data & customer feedback.
 Benz is moving away fro traditional background as a car maker
& positioning itself as a data-driven technology company.
 It competitors are just like to be Google & Apple as BMW or
Toyota.
 Production lines of the near future will be safer, faster and more
efficient, thanks to improved ability to collect and analyse data at
every step of the process.

Using AI to Explore Space & Distant worlds

 NASA will launch its next mission to Mars in 2020.
 So far 4 Mars Rover craft is on the surface of the Red planet.
 NASA- deep space probes- New Horizons mission to Pluto &
the Voyager mission to the outer reaches of Solar system .
 Challenges in space explorations is the limited amount of
bandwidth for communication back to earth- due to distance.
 Unmanned aircraft often be out of contact with humans for long
period of time.
 Ability to make autonomous decision about what information is
valuables to their earth bound operator is vital.
 Limited amount of power available to operate the spacecraft- far
from charging stations, further from Sun’s source of solar
energy, power usage must be predicted and monitored.
NASA

 Spacecraft- equipped with a large number of sensors to capture data/
information.
 Space is empty vacuum & planetary surfaces are comprised of lifeless,
inert matter, no different from that found on earth.
 Teaching space-faring machines to recognize the anomalous data is the
main purpose of AI work carried out by NASA.
 We want spacecraft to know what we expect to see & recognize when
observes something different.
 Smart systems monitors the power usage and determines which systems
are using more energy and what systems can be shut down.
 AI driven robots are also increasingly being used instead of human
beings.
 NASA currently uses a robotic system known as Robonaut 2 to assist
humans carrying out complex technical operations in the hazardous
situations of outer space.
 Robots are equipped with AI driven image recognition technology.
NASA-AI

 Earlier it to 24 minutes for information gathered by sensors to
reach earth & another 24 minutes for instruction based on
that information to be returned to the red planet.
 This delay is obviously far longer.
 With AI this information instant
 NASA is pioneering AI to help solve problems in outer space
as well as back home on earth.
 Space exploration generates huge volumes of data, and it is
far more efficient to use autonomous machines to work out
what is worth sending home & what can be discarded.
 Technology developed for space exploration often has
utilities back home on earth- and licensing them can help
fund the high cost of development & deployment in space.

Using AI to Tackle the Energy Transition

 Royal Dutch Shell is heavily investing in R & D of AI, which it
hopes will provide solutions to some of its most pressing
challenges.
 From meeting the demands of a transitioning energy market,
urgently in need of cleaner and more efficient power, to improving
safety on the forecourts of its service stations, AI is at the top of
the agenda.
 Current initiatives include deploying reinforcement Learning in its
exploration and drilling program, to reduce the cost of extracting
the gas that still drives a significant proportion of its revenues.
 Elsewhere across its global business, Shell is rolling out AI at its
public electric car charging stations, to manage the shifting
demand for power throughout a day.
 It has also installed computer vision-enabled cameras at service
stations, which are capable of detecting customers lighting
cigarettes – a severe hazard.
Shell

 Encouraging motorists to switch to an electric vehicle is seen as key to
reducing the Co2 emissions caused by humanity, and limiting their effect
on climate change.
 But it involves something of a chicken-and-egg problem. Motorists are put
off making the switch due to a lack of public charging terminals, and
forecourt operators may be slow to adopt them due to a lack of demand.
 Shell’s answer to this problem involves deploying AI to monitor and
predict the demand for terminals throughout the day, enabling power to be
supplied more efficiently.
 “If you think about it,” says Jeavons, “as a grid operator you’re operating
many, many electric charging posts … if all the cars plug in at the same
time and automatically start charging, you create a big load on the grid t,
by the way, can’t be filled by solar, because it’s 7 am or 8 am in the
morning.”
 “So, what we can do by understanding people’s charge profiles is we can
spread the load during the day, which basically means we can save the
consumer money.
Shell- Charging Efficiency

 Another initiative being trialed in Singapore and Thailand involves the use
of computer vision at service station forecourts. Computer vision –
cameras which can “think” and understand what they are filming – are
trained to watch out for the potential hazard of customers lighting
cigarettes in the vicinity of pumps and refueling vehicles.
 Camera data is processed by what is essentially the same technology
powering Google’s reverse image search, which allows the content of the
picture to be labeled and categorized.
 When an image is detected that matches what the algorithms “know”
(through training) is a person lighting a cigarette, alerts can be issued
allowing the forecourt staff to close down nearby pumps and reduce the
risk of fires or explosions.
 This relies on “edge processing,” with camera data being analyzed locally
to avoid the delay that would be inevitably caused by sending it to the
cloud and back before action could be taken. While it currently focuses on
spotting smokers, in the future the technology could also be trained to
detect other hazards such as reckless driving, criminal damage or theft.
Shell- Monitoring Forecourts

 Shell is involved in the entire oil and gas supply chain – from mining
raw hydrocarbons from the earth to refining them into fuel and
various other products, to retailing them to businesses and
individuals. AI is being rolled out or trialed at each step of this
process. Recent developments include the adoption
of Reinforcement Learning – a form of “semi-supervised” ML to
control its drilling equipment.
 While machine learning can work with either labeled data
(supervised learning) or unlabeled data (unsupervised learning),
reinforcement learning takes a middle-ground approach by
incorporating a reward system, dependent on the outcome of the
AI's "choices."
 As Jeavons says, “The key thing is you’re giving the [AI] agent the
autonomy to make the decision. But you’re providing input into the
model, so you’re providing reward or penalty functions on the basis
of what’s happening in the model, and how the model responds to
the set of conditions that you give it.”
Shell- Precision Drilling

 Algorithms designed to guide the drills as they move through a
subsurface are trained on historical data from Shell’s drilling records,
as well as information gathered from simulated exploration.
 It covers mechanical information from the drill bit, such as temperature
and pressures, as well as data on the subsurface from seismic
surveys.
 The result is that a Shell geosteerer – the human operator of the
drilling machine – is able to understand the environment more
accurately they are operating in, leading to faster results and less
wear, tear and damage to machinery.
 In many ways the challenge was similar to those faced by developers
working on self-driving cars – only instead of navigating hazards a
vehicle might encounter on the road, the drilling machinery must
autonomously adapt to changing conditions under the ground.
 “What GE expect is that this will probably never fully replace
geosteering as a discipline, but it will allow a single geosteerer to
support many more wells.”
Shell- Precision Drilling

 Shell employs AI solutions across its business with key to use
cases around meeting its energy transition targets.
 Drivers often cite lack of available charging stations as a
reason for continuing to choose fossil fuel – vehicles.
 Charging site owners don’t like to front the cost of installing
infrastructure before the user base is in place- but supply
infrastructure as a demand helps them to share some of the
risk with the shell.
 AI can be used to understand & predict energy demand at
recharging points, and can regulate supply to avoid adding
unnecessary strain at peak times.

Siemens
Using AI and Analytics To build the Internet of Trains

 Trains that are never late? But Siemens say that thanks to Big Data and
analytics it could soon be a reality.
 Train operators have responsibility for getting millions of people to where they
want to be every day. When journeys are delayed it causes not only
disruption to people’s schedules, but operators often face large financial
penalties.
 Siemens AG is one of the world’s largest providers of railway infrastructure,
serving rail operators in over 60 countries. Through harnessing Big Data,
sensors and predictive analytics they say they can now guarantee their
customers close to 100% reliability.
 It calls this the “Internet of Trains” – the on-rails segment of the wider “
Internet of Things” concept which describes how everyday objects of all
shapes and sizes can now be connected together online and given the ability
to communicate and capture data for analytic purposes.
 “We are helping our customers to get more value out of their assets,” Gerhard
Kreß, director of mobility data services at Siemens
 “Customers have invested a lot into their assets – trains, rails, signaling –
and our task is to help them get better returns.”
Siemens

 Sensors on an Internet of Trains system monitor everything from engine
temperature, to the open or closed state of doors, to vibrations on the
rails, and even image data from outside of the trains using cameras.
 Machine data on how systems are operating is also used, as are external
real-time datasets such as meteorological information.
 The data is used to achieve one of three ends. The first is improving asset
availability, and this includes the ability to carry out “predictive
maintenance” – which means making data-informed decisions about the
best time to replace and repair components, ideally before they break and
when the equipment would not otherwise be in use.
 Secondly, it can be used to increase energy efficiency, which has
environmental as well as financial benefits.
 This includes micro and macro-management – not only can the
performance of individual components be monitored to identify sources of
inefficiency, but the overall status of a rail network can be monitored to
ensure tracks are clear and all trains are where they should be. This
means they can travel at optimum speeds without having to worry about
what’s in front of them.
Siemens-Internet of things

 The third objective is to improve asset utilization. This basically means
the network is transporting as high a volume of passengers and cargo as
possible.
 “We use data from rail vehicles and the infrastructure. Classically these
are complex systems which have their own diagnostic elements so we
take everything we can learn from the on-board diagnostics and augment
that with sensor data and log file data,” says Kreß.
 As well as reliability, very positive results have been achieved in the area
of fault-spotting and predictive maintenance.
 “In the case of [German national train operator] Deutsche Bahn, we have
been monitoring the bearings, the gearboxes, motors and other elements
since October and we have not overlooked a single component failure – I
am very glad that we managed to pick them all up.”
 It is this level of reliability that has allowed Siemens to really put its
money where its mouth is, by offering uptime guarantees, meaning they
share the risks – as well as the rewards – with the train operators
themselves.
Siemens-Internet of things

 Reducing the delays & minimizing environmental impact are
the key drivers to the move towards a smart automated
system in rail networks.
 Sensor data can be overlaid with operational data such as
breakdown & maintenance reports to give a fuller
understanding of factors that cause delays when it is used to
train AI systems.
 Increasingly unstructured data such as visual data from
camera feeds will be a key ingredient in this mix. Image
recognition software helps make sense of this unstructured
data by turning it into information that machines can
understand and correlate with other data resources.

Tesla
Using AI to Build Intelligent cars

 Tesla has become a household name as a leader and pioneer in the
electric vehicle market, but it also manufactures and sells advanced
battery and solar panel technology.
 As a tech pioneer with a significant interest in the race to build and
market autonomous vehicles, it makes sense that today they would be
deeply interested in artificial intelligence. However, it was only this
month that the business’s billionaire founder and CEO Elon Musk
publicly announced it is working on its own AI hardware.
 This is definitely interesting if not exactly surprising. Musk, after all,
has been outspoken in his views about AI. As well as revolutionizing
almost every aspect of society, he has warned that it will cause
widespread job losses and possibly even start World war three.
 He is also a co-founder of OpenAI, a research organization dedicated
to ensuring that AI is developed and deployed in a safe, manageable
way so as to minimize any existential risk robots may one day pose to
humanity.
Tesla

 Not many details have yet been made public about Tesla’s new AI, though it is
believed it will process the “thinking” algorithms for the company’s Autopilot
software which currently gives Tesla vehicles limited (“level 2”) levels of
autonomous driving capability. Musk has said that he believes his cars will be fully
autonomous ( Level 5 Autonomous) by 2019.
 Tesla has been criticized by some for appearing over-eager to be first to bring
autonomous cars onto the roads, in the light of what is being seen as the first fatal
accident involving a car which was driving itself.
 But as a business decision, it is hoping its pushy tactics will pay off, with experts
concluding that the company has trumped its rivals in the data-gathering
department.
 All the vehicles Tesla have ever sold were built with the potential to one day
become self-driving, although this fact was not made public until 2014 when a free
upgrade was rolled out.
 This means the company has had a lot more sensors out on the roads gathering
data than most of its Detroit or Silicon Valley rivals, many of which are still at the
concept stage. Having just launched its first mass-market car, the Model 3 with a
price tag of $35,000, the company is expecting the number of its vehicles on the
road to increase by almost two thirds to around 650,000 in 2018 – and that’s a lot
of extra sensors.
Tesla

 In fact, all Tesla vehicles – whether or not they are Autopilot
enabled – send data directly to the cloud.
 A problem with the engine operation meaning that components
were occasionally overheating was diagnosed in 2014 by
monitoring this data and every vehicle was automatically
“repaired” by software patch thanks to this.
 Tesla effectively crowdsources its data from all of its vehicles as
well as their drivers, with internal as well as external sensors
which can pick up information about a driver’s hand placement
on the instruments and how they are operating them.
 As well as helping Tesla to refine its systems, this data holds
tremendous value in its own right. Researchers at McKinsey and
Co estimate that the market for vehicle-gathered data will
be worth $750 billion a year by 2030.
Tesla

 The data is used to generate highly data-dense maps showing everything
from the average increase in traffic speed over a stretch of road to the
location of hazards which cause drivers to take action.
 Machine learning in the cloud takes care of educating the entire fleet,
while at an individual car level, edge computing decides what action the
car needs to take right now.
 A third level of decision-making also exists, with cars able to form
networks with other Tesla vehicles nearby in order to share local
information and insights.
 In a near future scenario where autonomous cars are widespread, these
networks will most likely also interface with cars from other manufacturers
as well as other systems such as traffic cameras, road-based sensors or
mobile phones.
 Although details are scarce on the new AI technology that Tesla was
creating, its current AI – driven by a partnership with hardware
manufacturer Nvidia – is largely based on an unsupervised learning
model of machine learning.
Tesla

 The high level of road causalities we see each year shows that
human cognitive & motor skills are not ideally suited to the task of
piloting one-ton hunks of metal, at speed exceeding 100 Kmph.
 Machines can react far more quickly and safely and communicate
between themselves more effectively.
 Giving cars the ability to “learn” how to safely navigate is
dependent on gathering large volumes of data. This can be done
under simulated conditions but information gathered from the real
world is likely to contribute to better understanding of reality.
 There still exists a healthy scepticism in public opinion over the
safety of autonomous vehicles. Until there is enough data to
effectively counter this, politicians & legislators are likely to be
extremely cautious when it comes to creating a legislative
framework for their operation.

Volvo
Using Machine Learning to Build the World safest Cars

 Volvo cars-Swedish company – reputation of producing automobiles with
great record of safety.
 Cars are increasingly generating more and more data as they become
ever more connected and empowered by smart, Internet of Things
technology. The need to capitalize on this data is forcing auto
manufacturers to rethink their data strategies.
 Thanks to modern telemetry, vehicles have been gathering and
transmitting data on how they are used for several decades. But the real
explosion in data volume is down to customer data available from the
applications and services available to today’s motorists.
 The generally accepted definition of a “connected car” is a vehicle which
can access online information and use it to assist in the maintenance
and operation of the vehicle, as well as enhance the comfort and
convenience of drivers and passengers.
 Research suggests that by 2020, 75% of new cars shipped will fit this
definition. The glut of data that will come with connected cars presents
unprecedented opportunities for insight, but also significant challenges.
Volvo

 Identifying areas where analytics could provide the most benefit is part of
Wassen’s job. Since Volvo launched its first car with internet connectivity
in 1998, it has worked to evolve its data strategy, initially working on
combining warranty claim data with telemetry to predict when parts would
fail or when vehicles would need servicing.
 The growing complexity of this dataset, along with the richness of the
insights, prompted the business to vastly upscale its analytics technology,
and today, it works with Teradata to carry out predictive, machine-learning
driven analytics across petabyte scale datasets.
 Their Early Warning System analyses over one million events every week
to discern their relevance to breakdown and failure rates.
 As well as predicting failure and breakdown rates, the company has put
data to use in order to uphold its reputation as a maker of safe vehicles.
One pilot project launched last year and scheduled to run until 2017
involves 1,000 cars fitted with sensors to detect driving conditions. The
focus here is on monitoring the vehicles’ performance in hazardous
situations such as when roads are icy. Data is uploaded to the Volvo
Cloud and also shared with the Swedish highway authorities..
Volvo

 The third focus of Volvo’s analytic strategy is improving driver and
passenger convenience.
 Efforts here involve monitoring the use of applications and comfort
features to see what their customers are finding useful, and what is being
underused or ignored.
 This includes entertainment features like built-in connectivity with
streaming media services, as well as practical tools such as GPS, traffic
incident reporting, parking space location and weather information.
 “We are looking into what types of applications are being used and we
continuously measure this in order for us to understand what it is that the
customers want us to develop in the future,” says Mr.Wassen
 Of course, the next hot topic in the car world is autonomous vehicles, and
Volvo, unsurprisingly, see safety as the main beneficiary here.
 A National Highway Safety Administration report found last year that 90%
of US road accidents can be blamed on driver error. This helped firm up
Volvo’s belief that removing the driver from the equation will lead to the
largest ever reduction in accidents.
Volvo

 Has enabled to understand the faults & errors that occur within
its connected cars.
 Better anticipation of servicing and repair centres.
 Future of cars lies in autonomy and deep learning is the key to
making it a reality.
 Roll out will be gradual –cars becoming gradually more
autonomous before self-driving cars become the norm.
 Safety is considered to be one of the key benefits of autonomy
in cars- will reduce number of accidents caused by human error.

Mail your comments to
ramaddster@gmail.com
End of Part -5 ( Concluding Part)

Summary artificial intelligence in practice- part-5

More Related Content

What's hot (20)

Similar to Summary artificial intelligence in practice- part-5 (20)

More from GMR Group (20)

Recently uploaded (20)

Summary artificial intelligence in practice- part-5