Hire intelligence as your driver

No artificial flavors, preservatives, or colors. Simply pure artificial intelligence!

That's right, imagine having your car operated by what is considered (in some applications) as having a higher intelligence than humans. Having AI controlling at least one or more essential aspects of autonomous vehicles will, in the near future, become inevitably standard.

Artificial intelligence is the brain behind what allows a vehicle to be fully autonomous. As the vehicle is driving around and constantly collecting data, the AI in the vehicle is learning more and more. Further allowing the vehicle to understand what it should do in varying situations. What does this mean for regular people? Driving from point A to point B will be done for you automatically, and in a safer way.

But before you are ready to hop into one of these fully autonomous vehicles, or even what is currently commercially available now, a Tesla, wouldn't it be cool to know how this AI is working?

There are three key aspects of AI in autonomous vehicles that should be explained: Perception, Planning, and Control.

Perception

This stage is where the autonomous vehicle begins to understand what the raw information coming in through the sensors or V2V components (the information coming in from other vehicles) are. From here, it will continue on the planning stage.

Planning

After acquiring the information coming in, the autonomous vehicle will need to make certain decisions to achieve some higher-order goals. The planning stage is how the autonomous vehicle knows what to do in a situation, given the steps it was told and has learned. In other words, the processed information about the environment (e.g., the sensors/cameras and V2X components) with established policies and knowledge about how to navigate in the environment (e.g. do not run over pedestrians, slow down when approaching a stop sign, etc.) all come together and is sent to the control stage, where what was planned is executed.

Control

To keep it simple, control is the process of converting the intentions, goals, and plan derived from the planning system into actual actions using the hardware of the vehicle.

---

To help clarify exactly what these steps mean, here's an example:

Let's say a car is right in-front of red-light and needs to turn right. A little later, the red-light turns green. However, there is a pedestrian crossing. The cameras and sensors on the car are perceiving this, and then need to plan what to do.

At this point, the AI is intelligent enough to know that once a red-light turns green, it needs to accelerate. However, since there is a pedestrian, it knows not to run them over. Here in the planning step, even though the light is green, the AI plans that it must continue breaking until the pedestrian is out the way. Then it can stop breaking, and begin to accelerate.

Now that the vehicle has a plan, the AI then converts the plan into actual actions by controlling the actuators for breaking, accelerating, and steering.

In all, artificial intelligence is vital for the operation and safety of an autonomous vehicles. AI allows the vehicle to constantly learn what is occurring in the environment, and plan accordingly. This higher intelligence is constantly being improved, leading way for a much safer commute in the near future!