Self- Driving Cars

5 min readNov 23, 2023

Image Credit: https://ideas.ted.com/many-of-us-thought-wed-be-riding-in-ai-driven-cars-by-now-what-happened/

What are self-driving cars ?

You may have heard the term “self- driving car” a lot lately in the 21st century.Tesla’s Autopilot, General Motors Super Cruise,and Nissan ProPilot Assist just to name a few. But, what are “self- driving cars” really? Simply put, a self-driving car has these features, such as sensors or cameras, that allows it to drive without the help of a human driver. Hence, the name self-driving.

How does it work?

Okay, so we know what it does, but how do these features work together to yield a car that can drive itself? The answer is a bit complex, but it boils down to machine learning, an enormous amount of data from image recognition systems, and neural networks. According to bu.edu, Eshed Ohn-Bar , a Boston University Engineer, is programming the driverless cars to watch other cars on the road and predict how they will respond to the environment. They then will use that information in their own driving. Apart from supervised learning, the cars also have built in sensors, cameras, and lasers such as LIDAR ( Light Detection and Ranging) and radars. LIDAR shoots out pulses of narrow invisible infrared laser to create a picture of depth of the car’s environment; basically perceives spatial information. Currently, there is a piece on top of cars that spin around to do this, but with integrated photonics these detectors are being shrunk to less than tenth of a millimeter which will one day fit inside of a car’s light. That along with radars( Radio detection and ranging) that calculates distance and cameras that provide vision for the car, make up convolutional neural networks (CNN) which is the system used to make cars driverless.

Image Credit: https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcTamaGmhkjXPHu_sFomREcfSGp6dOinM2Vt-w:https://hips.hearstapps.com/hmg-prod/images/autonomous-car-driving-on-road-and-sensing-royalty-free-illustration-1585662413.jpg&usqp=CAU

Accuracy

Now the main factor that we want to know about self-driving cars is their accuracy. Is it safe? Well to do that autonomous cars need to be driven a certain number of miles without accident to gain a high confidence level. As driverless-future.com puts it , “ Simple formulas for binomial probability distributions can be used to calculate the number of miles which need to be driven without failure to reach a certain confidence level: 291 million miles need to be driven by a self-driving car without fatality to be able to claim with a 95% confidence level that self-driving cars are as reliable as human drivers…If we relax the required confidence level to 50%, then at least 67 million miles need to be driven without fatality before we can be confident that self-driving cars are safe.” So, basically more testing needs to be done and self-driving cars are going to change and get better over time before we can make a solid statement saying this is the accuracy of self-driving cars.

Pros

Image Credit: https://wp.technologyreview.com/wp-content/uploads/2018/10/m.i.tsmartcarchoicescolo01-9.jpg

So what can this AI machine do that benefits our society? Valienetemott.com lists a few which includes prevention of car crashes, traffic efficiency, better access and mode of transportation, and it’s environmentally friendly. As we know, about 94% of car crashes are caused due to human error. The systems and algorithms used in self-driving cars can reduce these accidents up to 90%. Plus, the cars have an ability to communicate with each other which will allow them to be more travel and traffic efficient which goes back into prevention of car crashes. People who cannot drive because of impairments or chose not to can use self-driving cars as an alternative while also being safe and having a reliable mode of transportation. Finally, autonomous cars look towards the future, so it will most likely be electric instead of burning harmful gasses into the environment. This can help with sustainability and creating a better world.

Cons

Image Credit: https://static01.nyt.com/images/2016/12/13/technology/how-self-driving-cars-work-1481671863640/how-self-driving-cars-work-1481671863640-superJumbo.png

Now that we know the pros, what are some unintentional consequences ? It turns out that hacking all the way to moral machine ethics can be a problem. Because these cars do use computer programming,there can be an issue with security. Skilled hackers could be able to wiggle their way into a car system and cause the car to malfunction. Additionally, there could be machine errors. Even though autonomous cars can prevent accidents from happening, it can not 100% eliminate the risk of accidents that are caused by machines due to error and bugs. The morals of the machine also need to be taken into consideration. If you have two negative outcomes like swivel left and hit someone or veer right and fall into a ditch which can cause the passengers inside the car to get injured, which would it pick? Furthermore, autonomous cars can cause people like taxi drivers, uber drivers, and bus drivers to lose their jobs. If those people don’t have a safety net, they could end up in a dire situation.

What programming languages are used?

It turns out that C or C++ is predominantly used when programming self-driving cars;however, other languages are being used as well.If we’re talking about one of the most famous of self driving cars, Tesslas, then the level of importance/use for programming languages( 1 being the most important/use 4 being less importance/use) would be:

C or C++

Python ( often in combination with C or C++)

Java and Javascript

Function based programming such as Ruby , Css, React, etc.

Examples of the code that would be used to program models in C++ is the kinematic model which can “ predict the state on the next time step by taking into account the current state and actuators as follows”( according to github.com):