TEST SYSTEM FOR TESTING A LIDAR DEVICE

Description

[0001] The invention relates to a test system for testing a LIDAR device, in particular a LIDAR device provided on a motor vehicle. Furthermore, the invention relates to a method for testing such a LIDAR device and a vehicle test bench for testing a vehicle.

[0002] As autonomous vehicles are developed, they are increasingly being equipped with various types of sensors. These sensors include cameras (e.g. 360° cameras) and radar systems designed to monitor the vehicle's surroundings. For autonomous vehicles, it is also important that they can not only detect obstacles, but also identify objects in the vehicle's surroundings. For this purpose, LIDAR (Light Detection And Ranging) devices are increasingly being installed in vehicles. These are devices with a method similar to radar for optical distance and speed measurement, but use laser beams instead of radio waves like radar. The laser beams are in the infrared range, particularly in the near infrared range (NIR), so that the visible light from the laser beams does not irritate other road users.

[0003] The functionality of LIDAR devices is well known, so a further description is not necessary here.

[0004] LIDAR systems operate with pulsed light beams (also referred to as light pulses or light pulses). The devices thus create an artificial image of the environment around the vehicle, from which the light beams are emitted. Depending on the sensor type, the data is generated as a point cloud or a segmented image.

[0005] For autonomous vehicles, the functionality of LIDAR devices is essential and safety-critical. Without the measurement results from the LIDAR devices that continuously scan the environment, the vehicle cannot move autonomously or can only do so with limitations.

[0006] Therefore, LIDAR devices require extensive testing. This is done, on the one hand, with system tests in a laboratory environment. On the other hand, the finished, operational vehicles are used in practical tests on real roads or test tracks to verify the functionality and reliability of the installed LIDAR devices.

[0007] To test LIDAR devices, corresponding situations can be simulated on test tracks, for example, using artificial vehicles or humans (manikins) that are moved on moving platforms relative to the vehicle under test or pulled by other vehicles. These tests are also very complex, especially since they can only be conducted at a relatively late stage of development, when the vehicle equipped with the respective LIDAR device is already largely developed.

[0008] Testing under real road conditions outside test tracks is also very complex and costly.

[0009] Gao Yanze et al., "A Programmable All-Optical Delay Array for Light Detection and Ranging Scene Generation," IEEE ACCESS, Vol. 7, July 29, 2019 (2019-07-29), pages 93489-93500, describes a system with a LIDAR scene generator that generates simulated optical LIDAR scenes under laboratory conditions. The simulated scenes can be fed directly into a LIDAR guidance system under test.

[0010] DE 10 2016 218 277 A1 describes a method for testing the functionality of a driver assistance system. The driver assistance system includes a sensor system for detecting the vehicle's surroundings. Instead of the normal surroundings, the sensor system can also detect an artificial target as a stimulus.

[0011] US Pat. No. 5,281,813 A discloses a test system for a laser rangefinder in which a detector is arranged near a laser source and detects the occurrence of a first laser beam. This allows a second laser signal to be triggered with a time delay to test the laser rangefinder.

[0012] A similar arrangement is also known from US 4 342 514 A.

[0013] The invention is based on the object of providing a test system for testing a LIDAR device with which the effort incurred during real road operation or operation on a test track can be reduced.

[0014] The object is achieved according to the invention by a test system for testing a LIDAR device according to independent claim 1, as well as by a corresponding test method for testing such a LIDAR device according to independent claim 12. The test system can advantageously be used in a vehicle test bench for testing a vehicle. Advantageous embodiments of the invention are specified in the dependent claims.

[0015] A test system for testing a LIDAR device is specified, comprising a controller for generating LIDAR information about an artificial LIDAR object based on a real object to be simulated for a test process, wherein the LIDAR information comprises at least two-dimensional outline information and depth information about a virtual distance of the artificial LIDAR object; and comprising a LIDAR image generation device for generating the artificial LIDAR object based on the LIDAR information; and comprising a LIDAR projection device for projecting the artificial LIDAR object onto a projection surface.

[0016] The test system makes it possible to recreate a real object as an artificial LIDAR object (avatar). To create a test environment, LIDAR information is defined and generated. For example, to test the LIDAR device, it can be specified that a vehicle ahead should be detected that is moving slower than the LIDAR device or a vehicle carrying the LIDAR device. For example, a truck ahead can be simulated or generated as an artificial LIDAR object in this way.

[0017] To generate the artificial LIDAR object, at least a two-dimensional outline (for a truck, for example, a rectangle) and depth or distance information are required. The distance information specifies, in particular, how far away the artificial LIDAR object should be from the LIDAR device in the selected test scenario.

[0018] Of course, scenarios can be designed using various artificial LIDAR objects for use as test scenarios. These include, in addition to vehicles driving ahead (cars, trucks, motorcycles, cyclists), static obstacles (guardrails, buildings, trees, structural obstacles), people (pedestrians at the side of the road, pedestrians crossing the road, zebra crossings, etc.), and so on. Since the LIDAR device is intended to support an autonomous vehicle, virtually any real-world object that might occur in real traffic can be simulated as artificial LIDAR objects.

[0019] The controller is capable of generating the artificial LIDAR object. The necessary data, which typically results from the specified test scenarios, can be appropriately processed by the controller to provide suitable artificial LIDAR objects.

[0020] Based on the artificial LIDAR objects, an artificial response image is simulated using the LIDAR image generation device and the LIDAR projection device. This artificial response image, corresponding to the artificial LIDAR object, corresponds to an image that the LIDAR device would perceive of the detected real object under real conditions, for example, in real traffic, and convert into an image.

[0021] According to the operating principle of LIDAR, an infrared laser beam would be emitted from the LIDAR device, and the reflected beam would be captured and analyzed. The light beams are pulsed in the form of light pulses and can cover larger areas, capturing both the area and depth information of the scanned environment.

[0022] The images captured by the receiver on the LIDAR device can be simulated in the test system by the LIDAR image generation device and the LIDAR projection device. The LIDAR projection device projects the artificial LIDAR object onto the projection surface, particularly using light pulses (infrared light pulses), from where the reflected light can be detected by the receiver on the LIDAR device. The receiver on the LIDAR device cannot detect that this is artificially generated light and not a reflected light pulse.

[0023] The depth information as a component of the LIDAR information can be at least one-dimensional and, in particular, based on the travel time of a light pulse, in particular an infrared light pulse (e.g., NIR light from a laser). The depth information, as well as the contour information, converts the distance and speed of the real object simulated as an artificial LIDAR object. Both pieces of information can be varied accordingly to simulate the movement of the artificial LIDAR object.

[0024] The projection surface can be designed in a variety of ways. It serves to "visualize" the LIDAR object, for example, on a screen or a ground glass. "Visualization" here means making it visible to an infrared receiver in the LIDAR device, but not to humans.

[0025] The depth information can be determined based on the actual (real) distance between the projection surface and the LIDAR device under test and based on the virtual distance of the artificial LIDAR object. Accordingly, the depth information must be based on a temporal component that is calculated or specified to simulate the (virtual) distance between the LIDAR device and the LIDAR object.

[0026] In particular, the travel time of the light pulses that capture the LIDAR object is determined as a temporal component, taking into account the speed of light. This calculation takes into account how long the light would take to travel from the LIDAR device's light transmitter (emitter) to the object to be detected and, after reflection from the object, back to the LIDAR device's light receiver (sensor) during real-world operation. Based on this time-of-flight, derived from real-world conditions, the controller simulates the transmission of each light pulse in such a way that it can be received by the light receiver during test operation of the LIDAR device, and thus for generating the artificial LIDAR object, even after this time has elapsed.The LIDAR device is thus fooled into believing that the artificial LIDAR object is also at the same distance as the real object, although it is actually projected onto the projection surface at a (much) shorter distance.

[0027] Since the distance between the LIDAR device and the projection surface is smaller than the (virtual) distance to be achieved between the LIDAR device and the artificial LIDAR object, the light pulse must be generated with a time delay that is determined accordingly by the controller.

[0028] As an example, a truck was given above, whose outline information can be simulated as a rectangular area. The depth information, for example, a distance of 60 m from the LIDAR device, must be determined by the travel time of the light pulse. Thus, from the target value of "60 m distance" specified for the test scenario, the time it would take for a light pulse to travel from the moment it is sent by the LIDAR device to the moment it is received by the LIDAR device is calculated. The LIDAR projection device is controlled in such a way that the IR light pulse depicting the artificial LIDAR object reaches the LIDAR device under test at the same time that the light reflected from the real truck to be simulated would also reach the LIDAR device again.

[0029] The LIDAR projection device can, in particular, comprise an IR laser device for projecting the LIDAR object as an infrared image onto the projection surface. The infrared image can be drawn either as a flat surface or using light pulses. In particular, an NIR laser is used for this purpose. IR LED light sources can also be used.

[0030] A LIDAR detection device can be provided to intercept and detect a light pulse emitted by the LIDAR device under test. It should be noted that the operating principle of the LIDAR device is to emit a light beam (IR light pulse) and, after reflection from the object to be detected, receive it back by the LIDAR device. For this purpose, the LIDAR device typically has a transmitter (emitter) and a receiver (sensor).

[0031] To ensure that the light pulse from the LIDAR device does not interfere with the simulated light beams or pulses in the test system according to the invention, the light pulses must be intercepted by the LIDAR device. In particular, no light pulses from the LIDAR device may reach the projection surface or even be reflected there. The projection surface itself in the test system will generally be located relatively close (a few meters) to the LIDAR device under test, while this is intended to simulate obstacles or objects at much greater distances (e.g., up to 300 m or more).

[0032] The light pulses emitted by the LIDAR device must be intercepted or dimmed accordingly.

[0033] At the same time, it is helpful for the test system's functionality to detect or recognize that a light pulse has been emitted by the LIDAR device. This detection can immediately trigger the controller to generate its own light pulse using the LIDAR image generation device and the LIDAR projection device to create the artificial LIDAR object described above.

[0034] In this way, the light pulse of the device under test (the LIDAR device under test) triggers the light pulse of the test system, which can subsequently be received by the receiver of the LIDAR device.

[0035] By detecting a light pulse emitted by the LIDAR device under test, the generation of a corresponding light pulse by the LIDAR projection device can be triggered. The LIDAR detection device and / or the controller can be configured to determine a location where the detected light pulse would have struck a real object, and the controller can be configured to generate an infrared light pulse with the aid of the LIDAR projection device. This pulse is reflected by the projection surface and is detectable by the LIDAR device under test, as if it were the light pulse emitted by the LIDAR device under test and reflected by the real object.

[0036] This enables the LIDAR detection device and, if applicable, the control system to identify which part of the projection surface a light pulse needs to be displayed in. If the test scenario does indeed call for an artificial LIDAR object at this location, the control system uses the LIDAR projection device to generate a corresponding light pulse, which can be received by the LIDAR device and subsequently allows for appropriate evaluation.

[0037] It is thus possible to determine at which location the light pulse would have hit the object if it had not been intercepted beforehand by the LIDAR detection device.

[0038] To enable subsequent correction of the triggering of the IR light pulse, the test system can include a device for reproducing customer-specific signal generation. This device is designed so that the necessary keys can be exchanged in encrypted form. This eliminates the need for the test system to have direct information about the signal processing of the LIDAR device. Instead, the control system in the test system is configured to adjust the time delay of the IR light pulse accordingly.

[0039] In the test system according to the invention according to independent claim 1, a video image generation device is provided for generating a visible video image of at least one real object on the projection surface; wherein the controller is configured to coordinate the generation of the video image and the generation of the IR image corresponding to the LIDAR object, such that the video image and the IR image are superimposed by the controller such that, in the resulting overall image, the displayed real object is superimposed with the LIDAR object.

[0040] In this process, a human-visible video image is superimposed on the IR image projected by the LIDAR projection device, which is invisible to humans. It is advantageous if the visible video image and the invisible IR image are synchronized.

[0041] The visible video image can also be received, in particular, by a camera installed in a vehicle being inspected, for example. The camera can also be coupled to the LIDAR device in the vehicle to detect obstacles, road conditions, etc., in the vehicle's surroundings.

[0042] For example, if the LIDAR object is intended to replicate the truck discussed above, the real object can be a video image of a truck filmed from behind, which the LIDAR device (virtually) approaches.

[0043] The LIDAR information generated by the LIDAR projection device can be precisely and synchronously overlaid with the visible video image.

[0044] An alignment device may be provided for aligning the visible video image and the IR image as well as the alignment of the camera and the LIDAR device in the test object (e.g. a vehicle to be tested) with the aid of a visible marker in the video image and a marker in the IR image.

[0045] It is in the nature of things that, for efficient testing, a precise overlay of the video image and the IR image is useful. However, this alignment cannot easily be performed by an operator, as they can only see the video image, not the IR image, with their eyes. For this purpose, it is possible to enhance the IR image, which is invisible to humans, with a human-visible marker. The alignment device can then overlay the corresponding marker with the IR image, so that the projected IR image also contains visible components that can be used as markers and aligned with the video image.

[0046] The alignment process can also be automated in this way.

[0047] The alignment device may include a marker image device for overlaying the IR image with the marker before the IR image is projected onto the projection surface. In this way, as already explained above, the IR light beam can be overlaid with a visible light beam.

[0048] In one variant, the test system is suitable for testing a stereo camera in addition to the LIDAR device, which is installed, for example, in a vehicle to be tested, but can also be provided independently, but in combination with the LIDAR device.

[0049] Accordingly, a stereo image device with two video image generation devices coupled to one another can be provided for generating a visible video image of at least one real object on the projection surface, wherein the video images each represent a view of the real object in such a way that, when combined, they represent a stereo image, the controller is designed to coordinate the generation of the video images and the generation of the IR image corresponding to the LIDAR object in such a way that the video images and the IR image can be superimposed by the controller in such a way that, in the resulting overall image, the represented real object is superimposed with the LIDAR object.

[0050] In this way, the test system can generate a stereo image, which can be fed alternately to the cameras by appropriately controlling the shutter devices arranged in front of the respective receivers or cameras of the stereo camera. Due to the different positions of the cameras, which can be arranged side by side, for example, a three-dimensional image can be reconstructed in a known manner.

[0051] The structure of such a stereo camera is known per se, so that a further description is not necessary at this point.

[0052] In the variant of the test system according to the invention described here, two different video images of a real object, each coordinated in terms of viewing angle and perspective, can be generated to also achieve the stereo effect. The test system designed in this way thus simulates stereo images, which can be alternately received or recorded by the cameras of the stereo camera under test by controlling the shutters.

[0053] In addition to the testing of a stereo LIDAR device described above, the video image can be generated three-dimensionally, e.g., using the two video image generation devices (projectors). This method can also be used to test a stereo camera on a vehicle, for example. The interaction between the LIDAR device under test and the stereo camera installed in the vehicle can also be tested. While the stereo camera evaluates the visible video image and uses this to detect objects, for example, the LIDAR device under test can determine additional information (e.g., distance, motion, speed).

[0054] Using the test system described above, it is possible to test a standard, commercially available LIDAR device, or one already installed in a vehicle and part of the vehicle's standard or optional equipment, without having to make any modifications to the LIDAR device. In particular, no structural or software modifications, or changes to the control system, are required. Rather, the LIDAR device can be used as it would be in real road traffic. For the LIDAR device, the artificial LIDAR object provided by the test system is indistinguishable from a real object.

[0055] The described test system can be used particularly advantageously in a vehicle test bench for testing a vehicle. Accordingly, a vehicle test bench for testing a vehicle is also specified, comprising a test system according to the above explanations for testing a LIDAR device installed on the vehicle, and a test system for testing the vehicle.

[0056] Vehicle test benches are known for a wide variety of purposes. They are used, for example, to test drivetrains (e.g., as a performance test bench), the vibration behavior of a vehicle, emissions tests, etc. The vehicles must complete a variety of test cycles, often legally prescribed ones.

[0057] According to the invention, a conventional vehicle test bench with a conventional test system can be expanded with the test system described above, thus also testing a LIDAR device installed on the vehicle. In this way, tests for the LIDAR device on the vehicle can be performed simultaneously during the test cycles conducted as part of development work, but often also due to legal regulations. This results in a significant increase in efficiency because multiple testing processes can be performed simultaneously.

[0058] Accordingly, it is a combination of the LIDAR test bench described above and a "traditional" vehicle test bench. This allows an entire vehicle to be tested in a single run, including a combination of LIDAR detection and vehicle behavior. If necessary, image recognition can also be tested by on-vehicle cameras based on the video projection described above. The vehicle test bench is thus capable of testing a wide range of functions of a test vehicle configured as an autonomous vehicle.

[0059] A method for testing or checking a LIDAR device is specified, comprising the following steps: Detecting a light pulse emitted by the LIDAR device under test; intercepting the light pulse before it reaches a projection surface arranged in the direction of the light pulse; generating an artificial response light pulse generated based on an artificial LIDAR object to simulate detection of the artificial LIDAR object by the LIDAR device under test; projecting the response light pulse onto the projection surface and reflecting the response light pulse from the projection surface; receiving the response light pulse reflected from the projection surface by the LIDAR device under test.

[0060] If necessary, the direction of the light pulse emitted by the LIDAR device can also be determined to increase accuracy.

[0061] The artificial LIDAR object serves as a virtual test object that is artificially generated.

[0062] These and other features and advantages are explained in more detail below using examples and the accompanying figures. They show: Fig. 1 a schematic representation of a test system for testing a LIDAR device; Fig. 2 a schematic structure of a LIDAR projection device; Fig. 3 a schematic diagram to explain how the test system works; Fig. 4 another illustration to explain how a test system works; Fig. 5 a schematic diagram of a test system for a stereo camera; and Fig. 6 the test system of Fig. 5 in a different operating state.

[0063] Fig. 1 shows a test system for testing a LIDAR device 1.

[0064] The LIDAR device 1 is constructed in a known manner and can, for example, be installed in a vehicle. It has an infrared scanner device (e.g., a NIR laser scanner or solid-state LIDAR) comprising a transmitter 2 (emitter) and a receiver 3 (receiver). In the test system, the LIDAR device serving as the test object can be tested with the vehicle or separately.

[0065] The operating principle of the LIDAR device 1 has already been explained above. Specifically, the transmitter 2 emits an infrared light pulse 4, which is reflected by a real object in real traffic and can subsequently be received and detected by the receiver 3. Of course, the transmitter 2 can also generate a plurality of light pulses 4 that cover an area to be detected by the LIDAR device 1.

[0066] Due to the propagation time of the light pulse 4 or the plurality of light pulses 4, a three-dimensional relief of the surroundings of the LIDAR device 1 can be captured using the receiver 3. In particular, the propagation time of the light pulses allows the determination of distance or depth information in order to be able to determine depth information as a third coordinate in addition to the area information that can also be detected by a radar, for example.

[0067] The test system comprises a controller 5 with an image generation device 6. The controller 5 is used to generate LIDAR information for a virtual or artificial LIDAR object (avatar) based on a real object to be simulated for a test scenario. For example, a test scenario may involve simulating a truck driving ahead. Based on the specified test scenario, the controller 5 is capable of generating the necessary data to simulate the truck driving ahead.

[0068] This data (LIDAR information) includes, in particular, two-dimensional outline information and depth information regarding the virtual distance of the artificial LIDAR object. For example, the truck can be represented by a two-dimensional rectangular area as outline information. Furthermore, the controller 5 can generate the associated depth information, which is intended to simulate the distance of the virtual truck from the LIDAR device 1 under test. The depth information can change in the process to also simulate the movement or speed of the truck.

[0069] Using the image generation device 6, it is possible to generate the artificial LIDAR object based on the LIDAR information generated by the controller 5 and define it as an infrared image. The image generation device 6 can be an integral part of the controller 5.

[0070] The controller 5 or the image generation device 6 controls a LIDAR projection device 7 with which the generated artificial LIDAR object can be projected onto a projection surface 8.

[0071] The projection device 7 can in particular be an infrared laser or an infrared LED that generates infrared light 9a (e.g. as an IR light pulse or as a plurality of IR light pulses) and projects the artificial LIDAR object in the form of an (invisible) infrared image 9 (IR image) onto the projection surface 8.

[0072] The LIDAR device 1 to be tested is positioned in front of the projection surface 8 and is able to receive the infrared image 9 from the projection surface 8 via its receiver 3.

[0073] This requires that all infrared light pulses 4 emitted by the transmitter 2 of the LIDAR device 1 be intercepted. For this purpose, a detection device 10 is arranged in front of the transmitter 2, which intercepts all light pulses 4 and, in particular, prevents the light pulses 4 from reaching the projection surface 8 and subsequently being detected by the receiver 3.

[0074] Furthermore, in one variant, the detection device 10 can be designed differently to detect the respective light pulse and, if applicable, its direction. In this way, by coupling the detection device 10 to the controller 5, the generation of a light pulse by the projection device 7 can be triggered. The light pulse 4 emitted by the transmitter 2 thus serves as a trigger for the generation of a light pulse by the projection device 7.

[0075] By appropriate coordination, the light pulse generated by the projection device 7 is emitted in the direction in which the respective current light pulse 4 from the transmitter 2 would also occur on an imaginary (virtual) object.

[0076] For example, since a light pulse that would travel 120 m from the transmitter 2 to a truck 60 m away and be reflected again in a real situation, a corresponding time would elapse before the reflected light pulse could be received again by the receiver 3. This delay is taken into account by the controller 5, which only generates the light pulse through the projection device 7 after a corresponding delay.

[0077] In this way, a planar infrared image 9 can be created.

[0078] In the test system, the light or laser pulses normally emitted by the LIDAR device installed in the vehicle are replaced by an artificial LIDAR signal. The controller 5 generates a virtual test object on the projection surface 8, which is generated in real time with respect to the parameters of distance, size, and, if applicable, movement (speed), and can be received by the receiver (LIDAR sensor).

[0079] A certain processing time is required to process the information that a light pulse 4 was detected by the detection device 10 and that a corresponding infrared light pulse 9a is to be generated by the projection device 7. This processing time should naturally be kept very short to minimize potential measurement errors. For example, it is in a range below 20 ns (e.g., approximately 10 ns, depending on the component design). This delay can lead to a deviation in the distance measurement. The corresponding error can be 3 m, for example, with a processing time of 10 ns. Accordingly, the distance to the virtual test object measured by the LIDAR device can subsequently be corrected taking the processing time into account.

[0080] To enable subsequent correction, the test system can include a device for reproducing customer-specific signal generation. This device is designed so that the necessary keys can be exchanged in encrypted form. This eliminates the need for the test system to have direct information about the signal processing of the LIDAR device.

[0081] In addition to the infrared or LIDAR projection device 7, a video projector 11 serving as a video image generation device is provided, which uses visible light 12a to project a visible video image 12 onto the projection surface 8, onto the same surface on which the infrared image 9 is projected. The video image 12 can also be artificially generated, but depict real driving situations in the various test scenarios.

[0082] Thus, the truck, which has already been discussed several times, can be made visible in the form of the video image 12 also for the camera installed in a vehicle as well as for an operator, while the infrared image 9 is, by its very nature, invisible to humans.

[0083] The video projector 11 can also be controlled directly by the controller 5 in order to coordinate the generation of the visible video image 12 and the generation of the infrared image 9.

[0084] It is also possible for the projection device 7 and the video projector 11 to be integrated into a common device, so that this device generates the infrared image 9 and the visible video image 12 equally.

[0085] The controller 5 can be configured to implement and simulate even more complex test scenarios. Accordingly, the controller 5 can generate data in real time that contains not only pure image data but also object data (e.g., corresponding to various object classes, such as "vehicles," "people," "traffic signs," "trees," etc.). This object list can be generated automatically according to the selected test scenarios. It can also be externally influenced by an operator of the test system for testing purposes.

[0086] Fig. 2 shows the LIDAR projection device 7 in greater detail.

[0087] It has an NIR (near infrared) light source 15 and a visible light source 16. The two light sources 15, 16 can each be controlled by the controller 5. The light beams from the two light sources 15, 16 are optically combined or superimposed in an optical combination device 17. Semi-transparent mirrors or similar devices can be used for this purpose. The result is a combined light beam consisting of visible light from the light source 16 and infrared light from the NIR light source 15.

[0088] The visible light from the light source 16 can be used, in particular, to create a visible marking on the projection surface 8. Using this marking, an operator can, for example, align the infrared image 9 and the video image 12 on the projection surface 8 (see Fig. 1 ).

[0089] For example, the light source 16 can generate visible light in the form of a suitable marking image with dots or crosses, which are generated on the projection surface 8 as needed.

[0090] An imaging device 18 is arranged adjacent to the light beam combining device 17. The imaging device 18, which can also be controlled by the controller 5 or the image generation device 6, can be formed, for example, by a DMD (Digital Mirror Device), an LCD (Liquid Crystal Display), or a DBD (Direct Beam Distribution) device. The imaging device 18 generates an infrared light pattern, which is subsequently projected onto the projection surface 8 using projection optics 19, where it generates the infrared image 9.

[0091] The infrared light is in particular pulsed light, whereby the depth information (third coordinate) is achieved by delaying the light pulses from the NIR light source 15, while the imaging device 18 essentially defines the two-dimensional outline information (position, size, etc.).

[0092] Fig. 3 shows an example of the interaction between the LIDAR device 1 to be tested and the test system in detail.

[0093] Accordingly, the transmitter 2 of the LIDAR device 1 emits the infrared light pulse 4, which is shielded by the detection device 10.

[0094] The detection device 10 detects that a light pulse 4 has been emitted and thus triggers the generation of a corresponding infrared light pulse 9a by the projection device 7 with the aid of the controller 5 or the image generation device 6.

[0095] The infrared light pulse 9a is projected onto a location 20 on the projection surface 8, which corresponds to the location 20 at which the light pulse 4 would impinge if it had not been shielded by the detection device 10. The corresponding direction of the infrared light pulse 4 is symbolized by a virtual line 21.

[0096] Accordingly, the generated infrared light pulse 9a strikes the location 20 and is reflected from there as an infrared light beam 9b so that it can be detected by the receiver 3 of the LIDAR device 1.

[0097] The temporal sequence is therefore as follows: Emission of an infrared light pulse 4 by the transmitter 2 - Detection of the light pulse 4 by the detection device 10 - Triggering of a corresponding infrared light pulse 9a by activating the projection device 7 via the controller 5 - Projecting of the infrared light pulse 9a onto the projection surface 8 - Receiving of the infrared light beam 9b reflected by the projection surface 8 by the receiver 3 - Determining the time between transmission by the transmitter 2 and reception by the receiver 3 - If necessary, determining the (virtual) distance of the artificial LIDAR object (virtual test object) generated at location 20.

[0098] Fig. 4 shows a similar situation as Fig. 3 , wherein the video projector 11 is additionally present, which projects the visible light 12a onto the projection surface 8 and thereby generates the video image 12.

[0099] For example, the previously discussed rear view of a truck traveling ahead can be imaged with visible light 12a at location 20. At the same time, an infrared light pulse 9a can be projected onto location 20 (here: the rear of the imaged truck) using projection device 7, which is detected by receiver 3.

[0100] The LIDAR device 1 is capable, in accordance with its intended function, of measuring the time between the emission of a light pulse by the transmitter 2 and the reception of the infrared light pulse 9a by the receiver 3 and of determining therefrom the (virtual) distance of the (virtual) object generated at the location 20.

[0101] The Fig. 5 und 6 show the structure of the test system as a stereo test system for testing a stereo camera device.

[0102] The devices of the LIDAR device already explained in detail above are no longer shown in the figures, so that only the differences or additions are explained.

[0103] Accordingly, the stereo camera device to be tested has two receivers or cameras 26a, 26b (e.g., a stereo camera in a test vehicle). A respective shutter 25a, 25b is arranged in front of each of the cameras 26a, 26b.

[0104] The video projector 11 is implemented in the form of two separate projection devices 11a and 11b, which alternately generate corresponding image information as visible light 12a. The image information from the projection device 11a is intended to be received by the camera 26a. Accordingly, in this case ( Fig. 5 ) the shutter 25a is opened while the shutter 25b is closed.

[0105] In the other case, the projection device 11b is activated, the image information of which is to be received as visible light 12a by the camera 26b. Accordingly, in this state ( Fig. 6 ) the shutter 25a in front of the camera 26a is closed, while the shutter 25b in front of the camera 26b is open, so that the light beam reflected from the projection surface 8 can reach the camera 26b.

[0106] To synchronize the interaction between the projection devices 11a, 11b, the cameras 26a, 26b (of the stereo camera under test), and the respective associated shutters 25a, 25b, the controller 5 has a synchronization device that ensures high-frequency switching. Depending on the design, the switching can be performed at a suitable frequency of 10 Hz or more.

[0107] As a result, a stereo image can be detected by the stereo camera under test, which can then determine and provide additional information in addition to the propagation times of the IR light pulses. This configuration enables the test system to also test a sensor fusion system installed in the test object (vehicle).

[0108] For the sake of clarity, the above description usually refers to a single light pulse. However, the actual function of a LIDAR device is based on the emission of a large number of light pulses, which ultimately create a planar infrared image, e.g., as a point cloud. Accordingly, the test

[0109] System capable of sending many light pulses at high speed with the help of the projection device 7 in order to generate a corresponding infrared image on the projection surface 8, which can be detected by the receiver 3.

Claims

1. A test system for testing a LIDAR device (1), comprising - a controller (5) for generating LIDAR information regarding an artificial LIDAR object based on a real object to be simulated for a test operation, wherein the LIDAR information comprises at least one piece of two-dimensional outline in-formation and one piece of depth information regarding a virtual distance of the artificial LIDAR object; - a LIDAR image generation device (6) for generating the artificial LIDAR object based on the LIDAR information; and comprising - a LIDAR projection device (7); characterized in that - the LIDAR projection device is configured to project the artificial LIDAR object (9) onto a projection surface (8); - a video image generation device (11) is provided for generating a visible video image (12) of at least one real object on the projection surface (8); and - the controller (5) is configured to coordinate the generation of the video image (12) and the generation of an IR image (9) corresponding to the artificial LIDAR object in such a manner that the video image (12) and the IR image (9) are superimposed by the controller (5) in such a manner that, in the resulting overall image, the displayed real object is superimposed on the LIDAR object.

2. The test system according to claim 1, wherein the depth information is determined based on - an actual distance between the projection surface (8) and the LIDAR device (1) to be tested; and based on - a virtual distance of the artificial LIDAR object.

3. The test system according to any one of the preceding claims, wherein the LIDAR projection device (7) includes an IR laser device (15), for projecting the artificial LIDAR object as an infrared image (9) onto the projection surface (8).

4. The test system according to any one of the preceding claims, wherein - a LIDAR detection device (10) is provided for intercepting and detecting a light beam (4) emitted by the LIDAR device (1) to be tested; and wherein - the light beam (4) is intercepted before it reaches the projection surface (8).

5. The test system according to any one of the preceding claims, wherein the detection of a light beam (4) emitted by the LIDAR device (1) to be tested triggers the generation of a corresponding light beam (9a) by the LIDAR projection device (7).

6. The test system according to claim 4 or according to claims 4 and 5, wherein - the LIDAR detection device (10) and / or the controller (5) are configured to determine a location (20) at which the detected light beam would have struck a real object; and wherein - the controller (5) is configured to generate, with the aid of the LIDAR projection device (7), an infrared light pulse (9a) which is reflected by the projection surface (8) and is detected by the LIDAR device (1) to be tested as if it were the light beam emitted by the LIDAR device (1) to be tested and reflected by the real object.

7. The test system according to any one of the preceding claims, wherein an alignment device is provided for aligning the visible video image (12) and the IR image (9) with the aid of a visible marker in the IR image (9).

8. The test system according to claim 7, wherein the alignment device includes a marker image device (16) for superimposing the marker on the IR image (9) before the IR image (9) is projected onto the projection surface (8).

9. The test system according to any one of the preceding claims, wherein - a stereo image device comprising two video image generation devices (11a, 11b) coupled to one another is provided for generating a visible video image (12) of at least one real object on the projection surface (8); - the video images (12) each represent a view of the real object in such a manner that they represent a stereo image by merging them; - the controller (5) is configured to coordinate the generation of the video images (12) and the generation of the IR image (9) corresponding to the artificial LIDAR object in such a manner that the video images (12) and the IR image (9) are superimposed by the controller (5) in such a manner that, in the resulting overall image, the displayed real object is superimposed on the LIDAR object.

10. The test system according to claim 9, wherein the test system is configured to test a stereo camera device with two cameras (26a, 26b), wherein - two shutter mechanisms (25a, 25b) which are assigned to the respective cameras (26a, 26b) on the stereo camera device to be tested and by means of which an incidence of light on the assigned cameras (26a, 26b) is blocked or unblocked as required; and wherein - the controller (5) includes a synchronization mechanism for synchronizing the two video image generation devices (11a, 11b) provided in the stereo image device with the shutter mechanisms (25a, 25b) in such a manner that a view of the real object can be alternately detected by one of the two cameras (26a, 26b).

11. A vehicle test bench for testing a vehicle, comprising - a test system according to any one of the preceding claims for testing a LIDAR device (1) present on the vehicle; and comprising - a test system for testing the vehicle; wherein - the test system is selected from the group of vehicle power train test system, performance test system, vibration test system, exhaust test system.

12. A method for testing a LIDAR device (1), comprising the steps of - detecting a light pulse (4) emitted by the LIDAR device (1) to be tested; - intercepting the light pulse (4) before it reaches a projection surface (8) arranged in the direction of the light pulse (4); - generating an artificial response light pulse (9a, 9b) generated based on an artificial LIDAR object to simulate detection of the artificial LIDAR object by the LIDAR device (1) to be tested, wherein the artificial LIDAR object is based on LIDAR information that has been generated based on a real object to be simulated for a test operation, wherein the LIDAR information comprises at least one piece of two-dimensional outline information and one piece of depth information regarding a virtual distance of the artificial LIDAR object; - projecting the response light pulse (9a) onto the projection surface (8) and reflecting the response light pulse (9b) from the projection surface (8); - receiving the response light pulse (9b) reflected from the projection surface (8) by the LIDAR device (1) to be tested; - generating a visible video image (12) of at least one real object on the projection surface (8); and - coordinating the generation of the video image (12) and the generation of an IR image (9) corresponding to the artificial LIDAR object in such a manner that the video image (12) and the IR image (9) are superimposed by the controller (5) in such a manner that, in the resulting overall image, the displayed real object is superimposed on the LIDAR object.