Automotive sensors: assistance systems' sense organs

5 min reading time
No modern car takes to the road without automotive sensors. There are engine speed sensors, and they’re also found in the air bags and on the wheels. Sensor technology is also important for detecting the environment around a vehicle, and thus for autonomous driving and automated parking functions. Radar, ultrasonic, cameras and soon lidar – read on for all you need to know about the sense organs of a modern car.

6 April 2021

Activate push notifications

Innovative mobility, exciting trends for the future and high RPMs: Subscribe now to get notified of new content.

Subscription successful.

Subscription failed.

If you need help follow the link for support.

1. Ultrasonic sensor / 2. Camera / 3. Radar
Each type of car sensor has its advantages. Together they complement each other perfectly.
Felix Modes

Development Engineer for Autonomous Driving Sensor Setups, BMW Group

Your car houses numerous automotive sensors – but they’re hidden as much as possible. Car sensors work unnoticed, making driving more comfortable, more efficient, and safer (➜ Read more: 6 safe driving tips). Read on as, with the help of BMW expert Felix Modes, we explain everything you need to know about car sensors and what role they play for automated driving and parking functions (➜ Read more: Overview of the main driver assistance systems), as well as for future autonomous cars.

No sensors, no car

Vehicles are equipped with complex technology for assistance systems. This includes numerous different types of sensor technology installed in various places and with different functions in the car. In addition to rain sensors, light sensors and similar comfort and safety functions, BMW R&D (research and development) and the R&D teams of other automakers and their partners are by and large working towards one goal: highly automated driving (➜ Read more: The value of time and autonomous driving).

What types of car sensors are installed?

In the assistance systems, sensor technology measures and evaluates what is happening in the immediate and extended surroundings of the car and detects any changes. Sensors are already employed in emergency braking assistants, night vision aids, traffic sign recognition, and of course in anti-lock braking systems (ABS) and dynamic stability control (DSC). In addition, certain BMW models have cruise control with braking, rear drive assist, an attention assistant, or a system that automatically detects necessary lane changes through active navigation. With the aid of BMW expert Felix Modes, we present in detail four types of automotive sensors and their functions: ultrasonic, cameras, automotive radar and lidar (also known as 3D laser scanners). Only when they are all working together do their strengths really come into their own.

Ultrasonic: robust and precise

Let’s start with ultrasonic. This piece of engineering technology is primarily used for parking functions, specifically reversing assistants, parking distance control and automated parking functions. Modes explains that this technology enables “very precise distance measurement over a short range.” Ultrasonic works by sending out pulses of sound and sensing how long these take to be reflected from an object. This robust sensor technology also has no problem operating in fog and darkness, although only at close range, i.e. up to around 30 feet (10 meters).

Cameras: high resolution makes them great for classifying things

Various types of camera are used in vehicles. Up to now, the best-known have been those used to make parking easier, such as the rear view camera. They operate with high aperture angles and have the advantage of capturing as large an area as possible (these are known as “fish-eye” cameras). For driving functions, on the other hand, cameras – and their ability to work with different focal lengths, from telephoto to wide-angle – are of great importance. These are located behind the windshield. Among all the different types of automotive sensors, cameras offer the advantage that they operate with a very high resolution and the image processing is able to accurately assess and differentiate between objects. As recognition is based on artificial intelligence (➜ Read more: AI design: AI gets creative), as Modes explains, “the image processing system must be taught the different types of object.”

Unlike other types of car sensor, camera sensor data can also be used to classify information such as traffic signs or the status of traffic lights. Their use is curtailed by the fact that they can be limited by environmental factors such as darkness and low-lying sun or a dirty lens. Distances and speeds can also only be estimated from the image data collection due to the principle of passive measurement – other types of sensor technology are better at this. All told, the sphere of application of cameras is broad – they can be used to identify the edge of the lane and support driver assistance functions such as lane departure warning and emergency braking functions, which react to vehicles, but also to pedestrians and cyclists.

Automotive radar: rain is no object

Most people associate radar (which stands for radio detection and ranging) with aviation technology. Yet this electromagnetic sensor technology is now also indispensable to automakers, who have been using it for the last 20 years. It’s used to measure speeds and distances. But how does radar technology work? Modes describes it as follows: “Radio waves are emitted, which scan surrounding objects. The echo of an object is analyzed and, if necessary, responded to.” When it comes to radar technology in cars, a distinction must also be made between two types, as BMW’s expert Modes explains. First up, there are short-range radars. They work with high aperture angles and low ranges (up to approx. 330 feet or 100 meters). Radar sensors of this kind are fitted as corner radars in the ends of the bumpers. They are necessary for the lane-change warning, the lane-change assist, and the intersection assistant. Long-range and full-range radar sensors, on the other hand, cover longer distances (up to approx. 820 feet or 250 meters) and provide information required by the emergency braking function and/or the adaptive cruise control (ACC).

This technology makes very precise distance measurements possible, and “influences such as rain or fog have hardly any effect,” explains Modes. Radar technology is less suitable for classifying objects. Depending on the design, there are various limitations in the technology's assessment of whether objects can be driven under or over. For example, imagine the system is faced with deciding if it is looking at the end of a traffic jam or a sign gantry. In such situations, the camera confirms whether emergency braking should be triggered.

Lidar: darkness is no problem

The three types of sensor technology described above can be found in BMW models today. Lidar is the next step. Like radar, lidar is also an acronym. It stands for light detection and ranging. Put simply, the system sends horizontally and vertically diffracted light pulses out into its surroundings. This scanning allows distances to be measured, taking fixed and moving objects into account. “The lidar sensor makes it possible to create a 3D map of the immediate surroundings, known as a ’point cloud’,” explains Modes. The big advantage of lidar is that it doesn’t depend on the ambient light and having to learn objects, as a camera system does. This means that lidar also responds safely to unknown objects. Thanks to its active light emission and very good resolution, lidar technology enables objects to be precisely classified even at night. These automotive sensors are still relatively expensive, but that will change as volumes increase. Lidar technology will be necessary to make the leap from level two of autonomous driving – where the driver monitors assistance systems – to level three, where the driver is able to hand over complete control to the car (➜ Read more: The 5 levels of autonomous driving).

Prerequisite for highly automated driving

So, as we can see, each of these automotive sensors has its strengths in a given area. Together they can create a perfect overall picture of the area around the vehicle – under any circumstances. And the overlapping of the systems ensures reliability and availability. Thanks to the progress of networking (➜ Read more: The connected car) and digitalization (➜ Read more: Smart cities), sensor data will become increasingly important. This means that automotive sensors are a crucial building block on the way to highly automated driving and eventually autonomous cars (➜ Read more: The development of self-driving cars). All of this technology is subordinated to one goal: the car of the future (➜ Read more: The car of the future) must get people to their destination safely.

Photos: BMW; Author: Nils Arnold; Illustrations: Madita O’Sullivan; Video: BMW