How important are sensors for autonomous vehicles?

Gerd, please tell us a little about yourself. What inspires you most about your job?

My name is Gerd Mittermaier and I am Global Vice President at MD. I have been responsible for the areas of product management, project management, research & development and patents since 2013. My professional background is in the semi-conductor industry where I acquired valuable experience in the development and implementation of technological innovations.

What I particularly love about my job is the fact that the automotive sector is so dynamic, and you have the opportunity to actively help shape modern solutions for transmitting data in vehicles. I focus on identifying technological trends at an early stage, developing new products and efficiently driving them forward from the initial concept to market maturity.

Close cooperation with experts from various specialist departments is what makes my job particularly exciting. I am really driven by the combination of innovation, strategic planning and technological progress. My goal is to develop sustainable solutions and to further strengthen MD as a leading technology partner for networked vehicles.

Let’s start with one of the most well-known sensors: the radar sensor.
Why exactly is this sensor so important for modern vehicles?

Radar sensors are essential for many safety functions in modern vehicles, such as Adaptive Cruise Control and Collision Warning. The sensors emit microwaves and measure how long it takes for these waves to be reflected by an object back to the sensor.

Based on this information, the vehicle is able to calculate the distance and speed of objects in its surroundings. There are three categories of radar sensors: long-range, mid-range and short-range, with measurement ranges of approx. 50 to 200 meters.

And what are the main advantages and disadvantages of this technology?

A major advantage is that radar sensors operate independently of lighting conditions. They are reliable day and night, in rain or fog. They also offer an extensive range, which makes them ideal for functions such as Active Emergency Stop Assist. On the other hand, the major disadvantage is that they offer lower resolution images than cameras, for example, which makes it more difficult to detect smaller objects.

Cameras are another frequently used technology in vehicles. What makes them so invaluable? 

Cameras offer very high resolution images and are particularly useful for detecting traffic signs, road markings and other details in the vehicle’s surroundings. They enable precise visual recognition which is of vital importance to many systems such as Lane Keeping Assist. Cameras are also relatively cheap and can be easily integrated into the vehicle.

But what are the challenges when using cameras?

Their main disadvantage is that they are dependent on lighting conditions. When it is dark or in unfavorable weather such as rain or fog, cameras are not as effective. Dirty lenses can also impair the function.
Another disadvantage is that cameras can generally only generate 2D images. To be able to depict a three-dimensional environment, the image data of at least two cameras has to be converted into a three-dimensional image. This is done by elaborate algorithms in high-performance vehicle computers, however, this always harbors a certain risk of misinterpretation.

In recent years, Lidar technology has emerged particularly in the development of autonomous vehicles. How does Lidar work?

Lidar (Light Detection and Ranging) operates by emitting laser beams that bounce off objects and return to the sensor. By measuring the time that it takes for the laser beam to return, Lidar can create a precise 3D image of the surroundings. This technology makes it possible to calculate distances to objects with a high level of accuracy.

The latest Lidar technology is even able to classify objects moving towards or away from the vehicle with the aid of the Doppler effect.
So it can measure the speed and respective distance to this object and interpret this in real time – which can be of crucial importance to autonomous driving.

What are the main advantages and disadvantages of Lidar and why is it so important to autonomous vehicles?

The main advantage of Lidar is its high precision and ability to deliver extremely detailed 3D images of the surrounding environment. It also operates in poor lighting conditions and therefore irrespectively of the time of day. That’s extremely important for autonomous vehicles which need to have an exact perception of their surroundings.

However, disadvantages include the high costs and limited efficiency in unfavorable weather conditions such as rain or fog.
Lidar is also unable to differentiate between colors and therefore cannot detect a difference between a plastic bag lying on the road and a bump in the road, for example.

Ultrasound sensors are frequently used for Parking Assist systems. What can these sensors do?

Ultrasound sensors emit sound waves and measure the time that these waves take to hit an object and be reflected back to the sensor. They are very well suited to close-range applications such as Park Assist or for detecting obstacles at a distance of approx. 15 cm to 5.5 m from the vehicle. 

What are the advantages and limitations of ultrasound sensors?

The greatest advantages are the low costs and the easy integration into the vehicle. They are ideal for short distances and offer good performance at low speeds. The disadvantage is that they only operate within a limited range and are not so effective at higher speeds or for longer distances.

Infrared sensors play a role in night vision and safety perception, what is their advantage compared to the other technologies?

Infrared sensors are particularly effective in detecting objects in complete darkness as they are based on thermal radiation. This technology can detect pedestrians or animals that are invisible to conventional cameras. So, they are very useful in situations where other sensors could fail.

Are there also disadvantages to using infrared sensors?

Yes, infrared sensors offer a lower resolution image than cameras and cannot capture details as precisely. They are also more expensive than many other sensors which somewhat restricts their distribution in vehicles and explains why they are also more likely to be found in the more expensive vehicle segment.

What challenges emerge from this multitude of sensors for data transmission in the vehicle?

Data transmission in vehicles is a complex challenge as sensors must be able to process huge amounts of data in real time. High bandwidths, low latency and reliable signal quality are crucial. Electromagnetic interference, temperature fluctuations and vibrations can impair communication. The integration of different sensors also requires unified standardization so that they can all work together flawlessly. Efficient data transmission is essential to optimize energy consumption and system performance. Only through innovative technologies can the requirements of modern vehicles be safely and reliably met, especially for autonomous driving.

Are hybrid solutions the future?  Which sensors should be combined in modern vehicles in order to achieve the best result?In principle, a combination of different sensor technologies is used to maximize the respective strengths and minimize the weaknesses. For example, many vehicles combine radar sensors with cameras to ensure both the range and the accuracy of details. Lidar is often also used in autonomous vehicles to create an even more exact 3D image of the surroundings.

It is important to network the different sensors together in order to develop a comprehensive and reliable driver assistance system.

This is known as sensor fusion or sensor data fusion, i.e. the process of combining sensor data in order to calculate detailed recognition and the most realistic possible interpretation of the surroundings in real time. The aim is to reduce the uncertainty of the individual sensors and increase the informational content to a maximum.

What role does MD ELEKTRONIK play in the future development of sensor technology?

MD ELEKTRONIK plays a key role when it comes to the reliable and long-lasting connection of sensor technology and control units in vehicles. In times where more and more sensors with ever increasing data rates and complex requirements are being integrated, our experience in the reliable transmission of signals and assembly is more sought-after than ever before.

We bring our expertise to the development process at an early stage – together with semi-conductor manufacturers, control unit developers and other partners. In the process, not only do we consider individual components, we also think about the entire technical connection – from chip to chip.

Our vertical integration offers particular added value: in addition to the development of components and processes, we also build our own production systems. This enables us to respond particularly flexibly to new sensor technology requirements and to offer our customers tailor-made solutions at the highest technical level.

At MD, we customize cable assembles and combine automotive-capable multi-core and coaxial cables in order to connect all types of sensors to the onboard network.

Gerd, many thanks for this interesting interview!