Self-Driving Cars - IoT - Udacity

Today, everything from phones and watches to refrigerators and thermostats are available in “smart” versions. You’ll often see all these devices bundled under the umbrella concept of something called “The Internet of Things,” or IoT. In simplest form, the term refers to what it sounds like: “things” connected to “the internet.”

IoT for Social Good

Despite the simple definition, IoT innovations are exciting, and in some cases even life-changing. For example, SweetSense makes smart sensors for hand-pumps in rural areas to monitor and optimize water flow. Propellor offers a smart inhaler system to help patients manage asthma. Bigbelly is transforming how cities address waste management challenges.

Everyday IoT

“Everyday” smart devices are also proliferating—speakers, cameras, heaters, doorbells, routers—everything from espresso makers to lawn sprinklers. These are fascinating, but much bigger changes are still to come. It’s one thing for a device on your wrist to count your steps, report them to the cloud, then send back actionable feedback. It’s another thing altogether when we’re talking about cars and trucks on the road.

The Evolution of Autonomous Transportation

“Smart” thinking is going big in the age of autonomy, and transportation is at the head of the innovation parade. Vehicles on the road have been getting technologically more sophisticated almost since their invention, but in the past decade the changes have been coming at ever-more dizzying rates, to the extent that the entire world is now engaged in conversations about what, exactly, smart transportation means in the IoT age.

A recent article in Business Insider offers a concise evolutionary summary:

“Step one was stop and go autopilot, which allows cars to drive themselves in traffic jams by analyzing the lane ahead of them and moving appropriately. Step two was the remote valet assistant, the ability to actually summon a car in a small space (such as a parking garage) through a smartphone, smartwatch, or key fob. Step three was highway autopilot with lane changing, which included blind spot technology to shift lanes.

Now, we’re starting to get into the truly exciting stage. Step four is cars that require a driver behind the wheel, but have an option for the driver to push a button to let the car drive itself. Step five will be totally driverless vehicles that do not require a driver behind the wheel, or even a steering wheel at all. BI Intelligence expects these cars to hit the market after 2020.”

Self-Driving Cars in the IoT Age

One way to think about IoT and self-driving cars it is to imagine every vehicle as a highly sophisticated mechanism for data collection—data about everything from road and weather conditions to driver behavior and brake, tire, and transmission performance. At first glance, this data is no different from what each of us, as individual drivers, gathers and stores in our minds as we travel. What is different, is that what we learn on our own, remains with us. Smart vehicles, on the other hand, send data to the cloud, and that information is then used to improve the software systems in ALL the connected vehicles.

Here’s a pretty staggering summary, from a recent article by Datafloq, of just how much information can be collected, using the example of a self-driving car from Google:

“The self-driving car from Google already is a true data creator. With all the sensors to enable the car to drive without a driver, it generates nearly 1 Gigabyte every second. It uses all that data to know where to drive and how fast to drive. It can even detect a new cigarette butt thrown on the ground and it then knows that a person might appear all of a sudden from behind a corner or car. 1 Gigabyte per second, imagine the amount of data that will create every year: On average, Americans drive 600 hours per year in their car. That equals 2.160.000 seconds or approximately 2 Petabytes of data per car per year.”

The data produced by self-driving cars can be understood as a kind of IoT equivalent of what Carl Jung called the “collective unconscious”; a sort of shared universal mind. In the case of self-driving cars—and in fact all “smart” devices—this collective mind is dynamic, and always being upgraded and optimized.

The implications are almost overwhelming when you think about the potential impacts on safety, efficiency, the environment, and more. Can you imagine real-time alerts about an accident ahead, from a car in front of you? How about self-driving cars “learning” from each other about fuel use optimization on mountain roads? Can you imagine what it would be like if all cars started up at the same time when a light goes from red to green? None of this is impossible.

Getting The Answers Right

There are concerns of course. Beyond all the obvious safety questions posed by self-driving cars—not to mention the ethical considerations—there is the issue of security. When a retail store’s systems get hacked, your credit card number may get used for fraudulent purchases. What happens if a self-driving car’s system gets hacked?

There is also privacy. Who owns all that data about how we drive, and what are they allowed to do with it?

Just as with any smart device in the IoT era, these questions will have to be answered for self-driving cars. But unlike with perhaps any other device, getting the answers right for self-driving cars can mean thousands, if not millions, of lives saved.


For more on this subject, see: Connected Cars That Care from IBM.


Ready to launch YOUR career in the field of Self-Driving Cars? We have two programs for you, depending on your experience level:

Intro to Self-Driving Cars Nanodegree Program

Self-Driving Car Engineer Nanodegree Program