has had seen a staggering increase in its population within the past 30 years.
The population doubled in the interval 1975-2018 reaching a total of 1.4
billion inhabitants. According to the World Population prospect, by 2024 India
will reach a population of 1.7 billion overtaking china as the worlds most
the population increase with the economic growth of the country and we arrive
at a total of 22,536,000 cars registered by the end of 2017. Due to the trend
of urban migration of the workforce a high percentage of these cars drive in
the urban centers of India. If we have a high concentration of cars in a
relatively small area of land, in this case a city, we soon realize that the emissions
of these cars are a major contributing factor to the problem of air quality.
has enabled modern society to import and export food and water from one side of
the globe to the other, yet the most important component to all life, air
cannot be transported. As humans, we can survive two days without water, three
weeks without food but only five minutes without air. An average adult takes
about 25000-30000 breaths per day but only eats 3 times a day and drinks about
2 liters of water each day.
order to keeps its future generations healthy, India needs to act now and
prevent further polluting its environment. The following proposal is an attempt
to act with currently available technology to manage the air pollution in major
urban centers. By no means, is this a solution which will solve the problem we
as a developing society face, but rather an attempt to control the damage done
and to prevent it from extending further.
main idea behind this proposal is continuous monitoring of air quality all over
the city. Sensors would be installed all over the city. Each sensor would then transmit
its data to a central station where the data will be analyzed. Since air
pollution cannot be modeled as a linear system, many more environmental
variables need to be take into consideration such as, air humidity, wind speed
and direction, temperature, precipitation etc.
the all the data is collected we fuse it together, similar to sensor fusion in
robotics for sate estimation, to obtain an end value. This value is then
automatically compared to a pre-calculated threshold value. If the threshold is
exceeded, an alarm is set off and the authorities are notified.
to magnitude of the overshot of the threshold value, a certain % of cars would
need to be withheld form the roads for a specific period of time. For instance,
the population would be notified two days in advance that for a period of two
weeks each day only cars with license plates containing randomly generated
alphanumerical combinations are allowed to drive.
approach is a novel one of course and would need the full cooperation of the
citizens to work. This however in reality most certainly would not be the case.
For this reason, CCTV cameras mounted on stoplights at intersections would
continuously monitor the license plates of cars during the restrictions. Anyone
caught driving a car with license plates which do not contain the allowed
alphanumerical combination would be issued a fine, which would be automatically
added to the property tax. Repeat offenders would see an exponential growth of
restriction would only apply to personal vehicles, and not to public transport.
In these days, public transportation would be supplemented with additional
Once enough data has been
gathered, the problem can be modeled as a machine learning task, namely as a
regression problem. The training of the neurons would commence in a supervised
learning manner, thus enabling the
algorithm to predict future air quality values giving the authorities more time