Future of Autonomous Vehicles (AVs) & the tech stack involved

The speedometer reads 55mph in a bold red digital font. You are slowly opening your eyes, which are still lazy from the long nap you just took in the back seat. Outside, the weather is pleasantly sunny. You see kayaks flowing on the lake, donut boats going around in circles.

Listening to your favorite book, you slowly sit up. The voice is almost distant, yet intimate & soft, enough to put you to sleep. You wonder if that’s a derogatory compliment for the author or the positive one for the narrator.

Shrugging it off, you casually remove the ear buds & place them in your bag. Now, trying to open your eyes fully, you gasp to see there is no driver! A rush of disorientation takes over & you start to panic.?

With the increasing heart rate, you try to make sense of this & suddenly realize that you are sitting in an autonomous vehicle, which you insisted on trying, even neglecting the exhaustion from hiking all the early morning!

Now fully awake, you realize that you are availing fully driverless ride hailing service for the first time in your life!

This might look like something straight out of some science fiction movie, but trust me this is already happening! The very scene depicted above is a common thing in Tempe, which is a city just east of Phoenix, in Arizona USA & the ride hailing service described is provided by Waymo!

Waymo is an American autonomous driving technology development company, which is a subsidiary of Alphabet Inc, the parent company of Google. Waymo operates a commercial self-driving taxi service in the greater Phoenix, Arizona area called "Waymo One", with Chandler, Arizona fully mapped.?

United States Department of Transportation (USDOT) predicts that AVs can reduce traffic deaths by a staggering 90%! According to reports, 42,915 people died in motor vehicle traffic crashes in 2021. With AVs, close to 39,000 of these lives could have been saved in a single year! Other benefits of AVs include increased lane capacity, decreased travel times, low emissions etc. Now, let’s dive deeper into the technology.

5 levels of automation in AVs

As the name suggests, autonomous vehicles are vehicles without drivers, so essentially the car is the driver. There are 5 levels of automation in AVs namely,

Level 0 - No Automation.

Our everyday car use.

Level 1 - Driver Assistance

Almost all cars come with this nowadays. Features are like cruise control, parking assist & lane keeping etc.

Level 2 - Partial Automation

In partial automation, the driver must have hands on the wheel, must be ready to take control. Automation helps to manage speed & steering.

Some examples are Tesla Autopilot, Volvo Group Pilot Assist, 奥迪 Traffic Jam Assist

Level 3 - Conditional Automation

So this is where things start getting interesting. Vehicles are capable of driving themselves, hands are off the wheel but still a driver is required behind the wheel. Conditions like limited access, speed limits need to be met for this.

Sensors act as eyes of the car & on board compute acts as brains to make decisions like speed, lane adherence, stop etc.

Level 4 - High Automation

This is the closest we are to automation & the prime example is Waymo AVs. It’s the expansion of partial automation to real world scenarios.

Level 5 - Full Automation

This is the future. No steering wheel, no pedals will be required. Vehicles will be self sufficient without any human intervention for driving!

Essential components of an AV

An AV is a fascinating combination of software, hardware, computing & sensors together with the application of AI & ML. Bosch , DENSO , AMD , 恩智浦半导体 , 大陆集团 , 英特尔 , Mobileye , 英伟达 & 谷歌 are some of the well known global players spearheading the AV revolution.

As the EVs accumulates kilometers, driving in the real world, they continuously learn from the past behaviors of pedestrians, other vehicles, environmental conditions & so on.

This essentially means the more AV drives, the more intelligent it becomes resulting in a more safe & efficient commute. Essential components of AVs are listed below.

Software

Many Driving Assistance System softwares are now being built. This is where decisions are being made. Software essentially dictates how the car will drive.

Interface - sensor fusion & mapping

This is the interface between hardware & software, the link between the two. This is where all the information from sensors is gathered & processed. Map info is used for routing & making calculated decisions based on ML.

Computing - CPU/GPU

This is the brains of the car where all the calculations & number crunching happens. This helps in real time decision making. Central processing unit or graphical processing, as the name suggests is the processing powerhouse behind the AVs.

Sensors - LIDAR, SONAR, RADAR, Cameras

These are the eyes & ears of the AV. These help the AV visualize & have a perception of the real world around it to make efficient decisions.

LIDAR - Light Detection and Ranging

It is a remote sensing method that uses light in the form of a pulsed laser to measure ranges. By targeting an object or a surface with a laser, this technique measures the time taken by the reflected light to return to the receiver to finally calculate the distance.

SONAR & RADAR - Sound Detection and Ranging & Radio Detection and Ranging

These respectively use sound & radio frequency to determine objects & ranges.

There are many challenges still plaguing the AV landscape like no clear regulation, lack of public support, trust issues, road/weather conditions? etc. but the future is most definitely here & I’m sure we will be able to tackle these challenges soon.

As we all evolve with new technologies, issues & resistances will be a common sight. AVs also will have their fair share of problems & adaptability issues. But a more connected & automated future is for sure right around the corner.?

Here’s hoping to share a ride in a driverless car, somewhere, someday. Till then, let’s keep our hands on the steering wheel & enjoy driving while we still can!?

Ohh & by the way, donut boats are a real thing found on the Tempe city lake, google it!