Full self-driving capability with                                                             Autopilot

Introduction

So, you were really looking forward to the day when you could just sit back and relax while your car drove you to work. But it looks like that day may never come.

You see, while cars have been accumulating automatic driver aids for years now, we still don't have a fully autonomous private car that can operate wherever we want it to. In fact, the Mercedes-Benz S-Class now has a model that can only drive around in one specific car park in Germany.

So don't give up on your dreams of being able to relax while you're driving just yet. We're getting closer and closer to making this a reality – it's just going to take a little longer than we thought.

Autopilot and Full Self-Driving Capability

You're probably familiar with autopilot features by now. Cars like the Tesla Model S and the Mercedes-Benz S-Class have had autopilot capabilities for a few years now. These features allow the car to take over some, or all, of the driving responsibilities under specific conditions.

But what if I told you that there was a model of the Mercedes-Benz S-Class that could drive itself around without any human input? That's right, the latest model of the S-Class has something called "full self-driving capability." This means that as long as you are in a specific car park in Germany, you can sit back and let the car do all the driving.

This is an impressive achievement, but it's important to keep in mind that we are still a long way from having fully autonomous private cars that can drive wherever we want them to. There are still a lot of technological and regulatory hurdles to overcome before we get there.

Autopilot and Full Self-Driving Capability Features

Your car can now park itself. You get in, tell it where to go, and it does the driving for you. The Mercedes-Benz S-Class has a model that can do this – as long as you are in one specific car park in Germany. It's an impressive achievement, but how far have we really come since New Scientist proclaimed 15 years ago that a fully driverless car "may not be far off"?

Cars have accumulated automatic driver aids, but a fully autonomous private car that can operate wherever you want to go is no more available today than it was 15 years ago. The big challenge for the automotive industry is to create a car that can drive anywhere, not just in a specific designated space.

Autopilot

That’s not to say nothing has been achieved. Cars can now park themselves, drive autonomously on motorways and handle complex urban environments with a combination of sensors, cameras, lasers and radar. Autopilot is a term used by some car makers to describe systems that can handle these tasks on their own — up to a point.

Autopilot is essentially an advanced version of cruise control, guiding the car in its lane and monitoring the speed limit. It takes into account other cars around it, obstacles on the road, weather conditions and traffic lights — but you still need to be holding the wheel. If you take your hands off the wheel for more than a few seconds then you will be warned and must take over control again quickly.

The autopilot function still requires human intervention in certain situations if it is unable to continue safely. However, it has been proven that this technology provides better safety than human driving as it can respond quicker in emergency situations by using its various sensors to detect danger ahead of time.

Full Self-Driving Capability

It's not like you're going to be buying a fully self-driving car anytime soon. Despite the fact that the Mercedes-Benz S-Class car can drive around unaided in a specific car park in Germany, it's not yet able to do this reliably and consistently everywhere.

There are still many challenges that need to be overcome before self-driving cars can become available for purchase. For instance, there needs to be an adequate amount of roadside sensors installed to help these cars navigate; regional and international regulations for autonomous vehicles need to be established; and, most importantly, their safety must be proven before they are released onto public roads.

So while full self-driving capabilities are certainly impressive and exciting, meaningful progress in this area is still years away.

Traffic-Aware Cruise Control

Traffic-aware cruise control (TACC) is a feature found in some cars that, in essence, gives the car the ability to maintain its speed and lane, autonomously. While this technology has been around for a few years now, it is still limited to highways and roads with clearly marked lanes. While this means that you won’t have to pay attention to your speed or lane position while on long stretches, it also means that you will still have to remain vigilant when navigating turns and sharp corners or changing lanes. In other words, while TACC provides a sense of autonomy on the highway, it is still far from offering up true autonomy when driving in an urban environment.

Auto Lane Change

Another feature of driving that is proving difficult to master is auto lane changing. This involves correctly judging another vehicle’s speed and direction, and making the appropriate calculations to switch lanes safely. Though advances have been made in this area, we are still some way from achieving a truly automated system.

Car manufacturers are also having to address the issue of “edge cases” – rare events that require split-second decisions for which no predetermined action can be set out in advance. If a child runs out in front of the car, for example, what does it do? 

What is certain is that we have come a long way since New Scientist first declared that a driverless car “may not be far off” 15 years ago. Nevertheless, it seems that a truly autonomous private vehicle that can operate wherever you want it to go is still some way away.

Navigate on Autopilot

But while you may be able to drive on autopilot in a car park, roadways are an altogether different challenge. Most self-driving cars have an array of sensors and cameras that need to process information from the environment they are in. For example, cameras can pick up lines painted on the road and use them to stay within the lanes, while radar can detect moving objects such as other cars and people.

However, autonomous vehicles must also be able to navigate around many environmental factors; everything from potholes or animals crossing the road to human behavior such as pedestrians walking out into traffic without warning or cutting off cars at intersections. This requires computer vision and machine learning technologies that do more than just identify objects in its environment but can also predict what those objects will do next.