A Brief Introduction To Self-Driving Cars

Sumeet Pathania
4 min readSep 17, 2019


Basics of self-driving cars

Self-Driving Cars

Just for a second close your eyes and imagine everyone on the road in self-driving cars. Well according to Morgan Stanley self-driving will be everywhere by 2026. In fact, many self-driving cars are already on the road.

Ok But Why Do We Need These Cars?

You including many others might not understand the importance of driverless cars. I mean why would anyone trust a vehicle to drive on its own! It doesn’t make sense. In fact, 63% of people in the US said they would never get in a self-driving car. However, data seems to suggest that autonomous cars(self-driving cars) may be safer than cars driven by humans. Especially, when most car accidents are caused by human error.

On average 1.25 million people die every year due to road crashes. That's 3287 deaths a day! Autonomous vehicles aim to reduce accidents and save more lives. In America alone, self-driving cars could save over 300,000 people per decade. Researchers estimate that driverless cars could, by mid-century, reduce traffic fatalities by up to 90 percent.

Cool! Now, How Do They Work?

Self-driving cars still have a long way to go. Currently, 5 key components help a self-driving car do its job.

  1. Computer Vision

When it comes to self-driving cars they drive similar to humans. Just like we humans, these cars need to see their environment as well and that exactly what computer vision is. Its how the car sees!

2. Sensor Fusion

Like I mentioned before, these cars are similar to humans. For example, we don’t just look at what's in front of us and keep driving. We have to analyze our surroundings and watch out for others that aren’t so courteous on the road. Similarly, self-driving cars use multiple sensors and radars like LiDAR and laser sensors to carefully examine their surroundings and then make decisions based off of that information.

3. Localization

Pretty much a fancy name for a very accurate GPS. Localization is the way the car know its locations with an error of less than 10 cm. Self-driving cars use maps because these maps help them figure out what is around them.

Human drivers can empathize with this. It’s much easier, and often much faster, to drive a route that you’re already familiar with because you know generally what to expect. Likewise, self-driving cars use maps that can tell them where to look(stop signs, intersections, lanes, etc). This helps the vehicle operate much more efficiently.

4. Path Planning

This component of autonomous vehicles plans the path of other vehicles and according directs its self. For example, if someone decides to change lanes the car sees it and adjusts accordingly(slow down, change lane, etc).

5. Control

After getting all the information, all the car has to do is steer its self in the right direction. And that's essentially the final component of a self-driving car.

Key points

It might have been hard to remember all of the components of a self-driving car. But think of it like this: Self-driving cars are very similar to humans. They need to see, understand their environment, know where they are, plan their path and finally steer.

That's Not All!

There are different levels of self-driving cars. Not all autonomous vehicles are completely independent. Hence, self-driving cars are split up into 5 categories/levels:

That’s it for this article. I hope you learned something new and cool.



Sumeet Pathania

Hi, I’m Sumeet and I’m a 19-year-old innovator. I am interested in emerging technologies such as Artificial Intelligence

Recommended from Medium


See more recommendations