AI on Roads: Saving the Time and People

Avanish Gupta
KasperTech
Published in
4 min readApr 20, 2021

For a long time, cities have been challenged for providing safe and fast intra-city transportation, especially the mega-cities like Mumbai, Delhi and Bangalore. There have been numerous experiments too, but none of them matches the competency and the self-improving nature of the artificial intelligence measures.

AI speeding up the transportation

Issues with manual detection

In recent times, we’ve seen many incidents of human and technical errors in managing road transport. Many of them could have been avoided with the use of appropriate technology. Even after having several rules and hundreds of quality checks by the automotive companies, there is a lot of scope for human error in handling vehicles as well as transport. And this is what paves the path of the entry of AI in this sector.

Data and Statistics

It’s horrible
It’s horrible

According to the National Crime Records Bureau(NCRB) of India, more than 154 thousand deaths were reported in India in roughly 440 thousand accidents in 2019. This is more than the population of several countries on the planet. The majority of accidents had one thing in common and that was over-speeding, which was a cause of death for more than 86 thousand people and injury to 271 thousand citizens. All this happens because India has just 2% of the land with a 16% population of the Earth. The challenge is to manage the resources efficiently and provide a better facility to the people.

Violations

AI can catch the traffic violations

I’ll say it is a satire on road safety in India that the Supreme Court has ordered the removal of the switch off button from headlights of automotive vehicles. There is no clear relationship between the frequency of accidents and weather/population/complexity of roads. They occur everywhere, from the narrow colonies in the cities to the widespread expressways, from a smoggy cloudy night to a clear sunny day, any accident can hit a vehicle out of the blue. The rules are broken everywhere too, leaving honest people to suffer more.

Effect on average speed and time of transportation

If we automate the transportation of a city, even to some extent, we will see a sharp spike in the average speed of vehicles throughout the city. For this, we need to keep track of the vehicles in the city, and their locations. With passing time, we can get the travel habits of the residents and the algorithms can keep modifying themselves in accordance with the travel habits of the residents.

Reduced accidents

It’s OK for monetary front, but why to lose lives?
It’s OK on the monetary front, but why lose lives?

In India, which has 1% of the roads of the world and 10% of the total road accidents, if we can save even 25–30% of those occurring in the cities, with an added advantage of the increased average speed of transportation, it will be a huge achievement from the humanitarian aspect.

Current Status

For the technologies like Artificial Intelligence and Machine Learning to operate at their full potential, the receptor city needs seamlessly fast internet connectivity, and some past records to start with. Apart from these, enormous computing power is required to organize and utilise the collected data from multiple sources. With 5G on the radar of researchers, and super-computing facilities increasing at expanding faster than ever, we are not very far from the days when AI will not just be in study projects but will be making our lives easier and faster.

No reason to refrain from it, then why not to use it?
No reason to avoid it, then why not use it?

Conclusion: To err is human and we cannot deny the fact that machines can free us from a lot of burdens, and they have done it well over the last century. It is the right time now to use the most advanced technologies in road safety. We’ve already lost a lot of human lives and we cannot afford more. Later we start, more lives we’d have lost. Because the algorithms get better with the experience, just like us, but with more reliance.

Vector Credits: Prince Patel

--

--