This article summarizes the new technological development
brought by Google – ‘Autonomous cars.’ The Google autonomous car project became
Waymo, a self-driving technology company with a mission to make it easy and
safe for people and things to move around. Cars today already include many
semi-autonomous features, like assisted parking and self-braking systems. And
completely autonomous vehicles—able to operate without human control—are
rapidly becoming more of a reality. The sensors in the car will be scanning
everything, and they are not affected by the driver’s mood (angry, tired,
happy). “Recent NHTSA research shows that approximately 94 percent of accidents
are caused by human error,” adds Alan Amici, a vice president of automotive
engineering at TE. “Cars with advanced safety features and eventually,
self-driving cars, can significantly reduce the number of collisions. The
impact of this innovation can be far-reaching, including reduced demand for
emergency response systems and reduced auto insurance and health care costs.”
It is estimated that 1.2million people are killed on the
world’s road every year. In America alone, 33,000 people are killed each year.
So, Google claims that it can reduce these deaths by 80% with the new
technology. The driverless car starts by understanding where it is in the world
by taking a map and sensor data and aligning the two. With the sensors, it
would be differentiating the other vehicles, pedestrians, etc. This car should
be able to guess other peoples’ moves and figure out how it should respond in a
moment. Something like what trajectory it should follow, how quickly it should
slow down or speed up. The car needs to understand where it is and roughly
where the other vehicles are which is a geometric understanding of the world.
These cars can access the data from the new traffic,
weather, surface conditions, construction, maps, adjacent vehicles and road
infrastructure. Information from these sources is used by the car for braking
or avoiding hazardous situations. “By
far the most complex part of self-driving cars, the decision-making of the
algorithms, must be able to handle a multitude of simple and complex driving
situations flawlessly,” Amici says. “The software used to implement these
algorithms must be robust and fault-tolerant.” These algorithms should reliably
catch the data from the sensors to make correct decisions on steering, braking,
speed and route guidance.
Google came up with lidar (a radar-like technology that uses
light instead of radio waves) sensor technology to bring a car without pedals
and steering wheels. Lidar builds 3D images of the surrounding landscapes.
Lidar contains some rotating stacked lasers that shoot at various angles. Every
layer is termed to be a channel and is made up of laser beams. The signal from
each circuit generates one contour line, and together these lines create a 3D
image of the object. Lidar also has a 360-degree vision and accurate depth
information. Unfortunately, this lidar technology is way too expensive.
A single autonomous car with all of its sensors, cameras,
and Lidar roughly could generate one gigabyte of data every second. That would
add up to 2 million GB of data per year per car assuming that on an average
American drives 600 hours a year. This data produced by a car can be classified
into three categories:
• Technical Data: It is the information
that comes from the car’s sensors and is analyzed by the car’s machine learning
• Community Data: data about traffic and
• Personal Data: riders’ personal preferences regarding
driving locations, indoor temperature, in-car entertainment, etc., all serving
to improve the user experience.
The autonomous car uses this data in making better decisions
while driving. Waymo seems to favor a higher-detail view of the world, with
Krafcik saying “The detail we capture is so high that, not only can we
detect pedestrians all around us, but we can tell which direction they’re
facing. This is incredibly important, as it helps us more accurately predict
where someone will walk next”.
The real hurdle for Waymo is not just in collecting the data
about the car’s surroundings, but to implement learning algorithms and improve
the driving capability, the more data it gathers and analyzes. For suppose, a
car can differentiate between a tube light and a potato chips and uses that
knowledge in future. It can conclude when a pedestrian is ready to cross the
street by observing behavior repeatedly. Deep learning algorithms can figure
out what is essential so that the car need not apply brakes every time a tiny
bird crosses its way. Hence, the car uses both predictive as well as
prescriptive models to improve its ability to recognize and react to its surroundings.
Apart from safety issues, Google’s self-driving car can
potentially create security and personal data issues. There might be a
situation where hackers can take control of the car. Intruders and advertisers
can take advantage of the exposed personal location data which is very
Big data creates an
opportunity for the self-driving cars in the future is to build a massively
connected network of vehicles, which will be able to interact with each other.
They could also communicate with roadside wireless sensor networks to sync up
information about traffic lights or accidents. It is firmly believed that
shortly, autonomous cars have the potential to become safer than human drivers
Data is transforming the way we think about transportation. It
has the potential to make driving more reliable and safe, by creating new opportunities
and insights. As data becomes the new paradigm in transportation, people who
can leverage the power of this data and build responsive, intelligent,
decision-making vehicles that move people and cargo around will be the winners!