LiDAR stands for Light Detection And Ranging or Light imaging, Detection And Ranging. The technology was made famous recently by a well-known smartphone manufacturer who used it to improve augmented reality content. LiDAR should also bring about the revolution of autonomous driving. But what does LiDAR technology actually do and what is the difference to radar?
A little excursion in advance: Why not use cameras?
After Elon Musk spoke out against the use of technologies other than cameras for his vehicles, there was a lot of criticism. A big problem here is that cameras do not provide sufficient information in poor visibility conditions, for example when driving at night. In addition, they cannot distinguish replicas of people, such as the "Attention, Children!” display, from real people. The advantage of the cameras, of course, is that they are cheaper. But LiDAR and radar sensors prove to be reliable sources even in poor visibility conditions.
Now to the main topic: what is LiDAR and what is the difference to radar?
LiDAR is related to radar technology, so let's start with an explanation. Radar stands for “Radio Detection And Ranging”. The biggest difference to LiDAR technology becomes apparent in the name: radar uses radio waves, while the LiDAR sensors use laser beams. Compared to light waves, radio waves have less absorption when they come into contact with objects, which means that they function over a longer distance. Radar systems are known primarily for military use, but also for measuring the landscape for mapping. The disadvantage of radar technology is that it has an inaccurate angular resolution compared to LiDAR, whose “point clouds” can display structures down to the smallest detail. For example, two objects that are located in the same distance in the opening angle will be displayed as one object because they are too close to each other to be distinguished by the radar.
And what about LiDAR?
LiDAR measures with bundled light in the form of laser beams. The advantage, as just mentioned, is the high resolution. LiDAR sensors send up to 150,000 laser signals per second. These signals are reflected by objects and the distance to the object can be determined very accurately using the returning signals. The time it takes for the pulse to return to the source (sensor) is measured. Thanks to this precise distance measurement, the use of LiDAR for autonomous driving has become indispensable.
There are different variants for recording the surroundings. With the so-called flash technology, a point cloud is created by illuminating the entire surface at the same time. The scanning LiDAR systems are more widely used and there are several types of these sensors. The types, which are based on the transit time principle, differ in how the individual transmitted laser pulses are deflected over the scene. For example, there are sensors that rotate so that the laser pulses are deflected in a full circle of 360°. To do this, several sensors have to be used so that their fields of view overlap.
Now we are continuing with another category that is also used by Ibeo Automotive Systems: solid-state sensors. These do not require any moving parts and are, therefore, much more compact. It is no coincidence that these sensors got their name. The pulses are deflected by a beam deflection unit, creating a larger field of view. The scanning is carried out by means of a structured deflection of the rays, which thus precisely capture the environment.
But what is “Truly Solid State”?
There is a lot of discussion in the LiDAR industry about what a “real” solid-state sensor is. In Ibeo's understanding, solid-state sensors do not contain moving parts. There are various options for covering the entire field of view (FOV).
1. The first is "Mechanical Scanning". The sensor or a mirror rotates in order to map a larger radius. This is not a solid state as there are obviously moving parts here.
2. The so-called "flashing", on the other hand, is a variant without moving parts. The entire FOV is illuminated with a "flash". This requires a lot of light and, therefore, a lot of optical power. That's a real solid-state approach.
3. MEMS  mirror technology is controversial, at least with regard to the solid-state status. A mirror moves minimally on a horizontal plane. Since there is still movement, it is not completely solid-state in our opinion.
4. With “Spectral deflection”, which is used in combination with a fiber sensor, different wavelengths are emitted. These waves are in the infrared range  and are emitted onto a prism, so that a larger FOV can be imaged. However, due to the pure wavelength consideration, this procedure is one-dimensional. It's a true solid-state approach, though.
5. Optical Phased Arrays (OPA) describes a control of the phase and amplitude of light waves in so-called "wave guides". This shape is also a solid-state concept. In order to get a two-dimensional representation, this technology could be combined with the spectral deflection in the future.
The advantage of a “real” solid-state sensor is that it requires less space due to the lack of moving parts, the design is simpler, and the price is, therefore, lower. This makes our sensors very compact, and cost-effective.
The solution: ibeoNEXT
Ibeo decided on the version of the "Sequential Flash LiDAR" - an improved version of the flashing technology: the ibeoNEXT. Less light is lost, and the optical power can be used more effectively. This creates a high-resolution image or a 3D point cloud. The module itself is so compact that it fits onto a credit card.
Climate change is ongoing. The events sector also contributes to this development because it is closely linked to the travel and hotel industries and, therefore, has a wide reach. In this post, Ibeo has analyzed how environmentally friendly events, in which we participated, are.
Although the term "autonomous driving" is used so frequently, many people are not even aware of what is actually behind the designation. In this article, we highlight SAE's five levels of automated driving and show where we currently are on this scale.
To this day women are still underrepresented in STEM subjects. This article adressed one of the reasons for the complex issue. Furthermore, Julia Nitsch, one of Ibeos algorithm developers, will be introduced to you.
Subscribe now and keep up-to-date.
By clicking ‘SIGNUP’ you consent to your personal data entered being used to send you information on relevant topics and specific publications of Ibeo. You have the right to withdraw your consent at any time. The withdrawal of consent will not affect the lawfulness of processing based on consent before its withdrawal.