If autonomous technology doesn’t excite you, then we’re not sure what will. It’s an exciting time to be alive and it won’t be long before the automotive world is turned completely on its head.
Find out what lies at the heart of autonomous technology, how it has been made possible, and what you can expect on the market today.
What is an autonomous car?
Also known as a driverless car or self-driving car, an autonomous car is one that makes use of different cameras, sensors, and intelligence to travel from one place to another without the need for human operation, sounds great, doesn’t it?
It gets a bit more technical when we’re talking about the different levels of autonomy and for a car to be completely autonomous. It has to be able to self-navigate to a destination using streets and roads that are yet to be modified, scary, or what! If, however, humans are required to be present, then obviously the car wouldn’t be classed as being fully autonomous.
How do driverless cars work?
It’s probably not news to you that autonomous cars are really intelligent, some might say that they’re even smarter than us humans. The way they work in getting from A to B is by using artificial intelligence. When it comes to programming, huge quantities of image recognitions data are installed and the vehicle will use this along with neural networks and machine learning, all sounds very technical.
In simple terms, neural works in the same way that our human brain does and it enables machine learning through data patterns. The data is processed is collected from your cameras on the streets and installed into our autonomous cars. It’s very clever if you think about it. This neural network helps to identify many different things like pedestrians, trees, traffic lights, and other vehicles. Anything that would be considered to be an obstacle when driving, picks up.
Now compare this to your traditional car, the control of the car lies with the human driver, who uses their eyes to see objects process the information, which then passes through the brain and into the driver's arms and legs, etc, which then carry out the appropriate tasks. To take the human out of the equation here sounds madness but something that we’re becoming more and more familiar with.
Years ago we couldn’t even begin to imagine letting our car be controlled by a computer, but that just shows us how advanced technology is these days and where the future is going.
That brings us to the beasty part of the autonomous car – the sensors. It’s the sensors in an autonomous car that basically steal the show and take credit for performing the function of the eyes. They detect and observe the objects in the surroundings, picking up on every little detail. This information is then passed to the car’s computer, which transmits to the car’s actuators, which would be similar to a human’s arms and legs. The actuators carry out the important task of braking and accelerating.
There are different kinds of sensors that are used by autonomous cars today; each of them with its own unique feature. Let’s take a glance at what these are below:
Cameras in terms of autonomous driving actually work in the same way our human vision does believe it or not. They actually use the same technology similar to the one found in our everyday camera that we use today. Your standard autonomous car comes with an average of 8 cameras fitted, sounds like a lot we know, but these cameras do a hell of a lot to scan the environments and seek everything out from different angles.
Having this many cameras, all around the front, side and back give the car good depth perception, which is essential for autonomous driving.
Lidar stands for Light Detection and Ranging and it's basically a remote sensing method that used light (laser beams) to measure ranges to the earth. We know this sounds fascinating and well weird but check this out.
So, a lidar essentially sends out millions of laser light signals per second and measures how long it takes for them to bounce back. It's very much to do with the speed of light in this part, and the time it takes for a beam of light to travel and hit an object and bounce back. All we care about though is the beam travelling to the object and back to the lidar.
What this means is that a lidar makes it possible to create a high-resolution picture of a car’s surroundings, and in all directions, if it is placed in the right spot, for example, the roof of the car, etc, where it can keep rotating to scan the surroundings. The great thing about this is that it can continue to be precise even in the dark too as the sensors are their own light source.
Radar is short for Radio Detection and Ranging, which is also a remote sensing method. The only difference is that instead of using light, it uses radio waves/frequencies to measure ranges to the earth.
In other words, what this basically means is that radars use radio waves to detect objects and determine their range, angle, and velocity.
4. Ultrasonic sensor
The ultrasonic sensor works similar to the radars and lidars, except for one thing: sound waves. The sensor head emits an ultrasonic sound wave and receives the wave when reflected. There’s not much more to it than that really, other than all the maths behind it and calculating the distance to the object and the time taken for the sound wave to hit and come back, but we won’t go too much into that.
5. Inertial measurement unit
Sounds fancy, doesn’t it? This is an electronic device that calculates and reports an object’s specific force and magnetic field surrounding the body of the car.
What this measurement does, in theory, is it can help to detect if the car is already moving or not and if so, in what direction. Its thanks to accelerometers and gyroscopes from the inertial measurement unit that this detection can take place. These two detect direction and rotational motion.
The future of driving
There are so many key players in the autonomous driving industry today and we can expect this to keep expanding as the years go by. The developments of self-driving cars started just before 2000, and other technical advancements soon followed.
Tesla and Google appear to be taking the industry by storm so far and they aim to collect and process enough data to create a car that can drive itself eventually. With others such as Audi and Mercedes yet to make an appearance, we can expect that as technology advances, it won’t be long before they put themselves on the autonomous map.
2010 saw the arrival of more assistance technologies like lane-centring assistance, which was followed by self-parking and adaptive cruise control in 2016. Come 2018, and although there are no fully automated cars available to the public yet, automated driving is paving its way onto the scene, and we can’t wait!