Fraunhofer Institute for Microelectronic Circuits and Systems
Fraunhofer Institute for Microelectronic Circuits and Systems
Autonomous driving concept illustration - 3d rendering showing a top down view of lidar sensor use

© Mopic/261809839/stock.adobe.com

ATLAS - Fraunhofer IMS LiDAR Target Emulator

Comprehensive and comparable real tests with virtual scenarios for object and environment detection are revolutionizing driver assistance systems

With the progressing development of automated functions for automobiles, the need for time-saving and comparable verification tests of the used measurement sensor technology are increasing. To be able to test environmental sensor technology for the detection of humans and objects more reliably and extensive, Fraunhofer IMS is developing the novel method of the »LiDAR Target Emulator« in the form of the system »ATLAS« (Automatically Testing of LiDAR Applicative Situations).

Figure 1: ATLAS as a measurement method for fast and comparable verification of LiDAR sensor technology at Fraunhofer IMS
© Fraunhofer IMS
Figure 1: ATLAS as a measurement method for fast and comparable verification of LiDAR sensor technology at Fraunhofer IMS

LiDAR Target Emulator

With a LiDAR Target Emulator, real LiDAR sensors (LiDAR: Light Detection and Ranging) can be virtually tested with arbitrary environmental scenarios.

The emitted light signals of the sensor are manipulated in such a way that arbitrary objects with predefined distances can be projected onto the sensor. Real LiDAR sensors can be tested using a LiDAR target emulator with arbitrary and repeatable scenarios. This requires a measurement screen and an intelligent algorithm to virtually place objects, such as the two vehicles shown in the LiDAR sensor's field of view (Fig. 1).

Figure 2: Basic overview of function and application of ATLAS
© Fraunhofer IMS
Figure 2: Basic overview of function and application of ATLAS

In addition to the LiDAR sensor (device-under-test, DUT), a separate measurement screen is required to perform the measurement, which is placed in the sensor's field of view. Specific laser signals, such as wavelength, pulse shape or spatial distribution, are received by the measurement screen and processed with a corresponding algorithm. These are then returned as response signals from virtual objects. This measurement method is also referred to as over-the-air measurement, since there is no electrical connection between the LiDAR sensor and the test equipment.

The corresponding LiDAR system may already be integrated in the vehicle and incorporated in a hardware-in-the-loop (HiL) test bench, or it may be placed as a stand-alone device in front of the measurement screen. In this way, an end-of-line (EoL) test can be performed after the LiDAR has been manufactured.

Unique tests under reproducible conditions with LiDAR target emulators

The method combines the previous verification methods for driver assistance systems, real tests and virtual environment simulation in a single measurement. This combination can save an enormous amount of time while increasing the scope of testing.

The biggest advantage of the LiDAR Target Emulator is that the measurement conditions are always the same due to predefined virtual scenarios. The measurement results are thus comparable both for a sensor that has been measured several times and between different sensor solutions. Environmental conditions in particular, such as sunshine, rain or fog, are important test scenarios for LiDAR sensors and can hardly be reproduced so far in real tests or only with a great deal of effort. With the LiDAR Target Emulator these weather influences are the same for all test specimens. In addition, this measurement setup can generate a large amount of valuable test data that is required for further training and optimization of the sensor.

Figure 3: Potential properties of ATLAS
© Fraunhofer IMS
Figure 3: Potential properties of ATLAS

The LiDAR target emulator method ATLAS from Fraunhofer IMS offers all the mentioned potentials for a comparable, standardized and time-saving verification of LiDAR-based driver assistance systems. The method is based on a novel concept of virtual representation of objects in real space. The  graphic above shows the potential, which forms the basis of ATLAS (Fig. 3).

Recent Publications

  • Publication S. Grollius, M. Ligges, J. Ruskowski and A. Grabmaier, »Concept of an Automotive LiDAR Target Simulator for Direct Time-of-Flight LiDAR«, in IEEE Transactions on Intelligent Vehicles, doi: 10.1109/TIV.2021.3128808.; November 2021: description of the conceptual relationships of the LiDAR target emulator ATLAS.
  • S. Grollius, J. Ruskowski, M. Ligges, and A. Grabmaier, »LiDAR Target Simulator considering background light and mutual interference,« in AutoTest, Stuttgart, 2022.