DE102022202206A1 - LiDAR system and method for operating a LiDAR system - Google Patents

Abstract

A LiDAR system (1) is disclosed which is set up to scan an environment of the LiDAR system (1) with a light beam (4) in order to obtain depth information about the environment, the LiDAR system (1) further to this is set up to use the light beam (4) to gain information about an object (5) that is in the environment. The LiDAR system (1) is set up to use the light beam (4) to obtain information about the object (5), which is located in an area of the environment to which there is no direct line of sight for the LiDAR system (1). . A corresponding method for operating a LiDAR system (1) is also disclosed.

Description

The present invention relates to a LiDAR system that is set up to scan an environment of the LiDAR system with a light beam in order to obtain depth information about the environment, the LiDAR system also being set up to use the light beam to obtain information about an object that is in the environment.

The present invention further relates to a method for operating a LiDAR system, comprising the step of scanning an environment of the LiDAR system using a light beam of the LiDAR system.

State of the art

In principle, non-line-of-sight imaging is state-of-the-art. These are technologies that, using different signal propagation paths of a signal (the most impressive are diffuse reflections), can capture the image of the hidden area where the object is located without a direct line of sight. However, such known systems are still limited in terms of application. The signals have to be very clean, the conditions have to be checked often, the propagation paths have to be partially learned, highly sensitive sensors are used, etc. The technology has not yet left the research laboratories for civilian use.

Frequency Modulated Continuous Wave LiDAR (FMCW-LiDAR) is also state of the art. The frequency of the LiDAR system can, for example, be modulated as a frequency ramp over time. The principle is known from radar. When a signal is received, the frequency shift of the received signal compared to the current frequency is made up of the transit time effect (frequency shift while the beam has traveled back and forth) and the Doppler effect (frequency shift due to movement of the object). Multiple ramps allow to unambiguously determine distance and speed for multiple objects in the environment.

The DE 10 2014 207 896 A1 discloses a '3D coarse laser scanner'. A module for measuring an object is proposed there, the module being configured to generate a primary beam, the module having a scanning mirror structure, the scanning mirror structure being configured to deflect the primary beam in such a way that a scanning movement is carried out by the primary beam, the Module is configured in such a way that a secondary signal can be detected if, in a deflection position of the scanning mirror structure, the secondary signal is generated by interaction of the primary beam with the object, the module being configured to generate a locating signal depending on the deflection position of the scanning mirror structure, the module has a sensor arrangement for generating a sensor signal for position-based measurement of the object.

The DE 11 2019 000 057 T5 discloses a method for vehicle-to-pedestrian collision avoidance. The method uses LTE radio signals to determine spatiotemporal positioning to alert a pedestrian and vehicle via an emergency signal when they reach a proximity threshold limit. In this method, the vehicle-to-pedestrian proximity threshold is preferably no more than 10 meters.

In the DE 10 2014 219 165 A1 a system for real-time detection, identification, location and tracking of vulnerable road users in a region of interest by a radio frequency based system embedded in the vehicle and on the vulnerable road user, under LOS (line-of-sight) and NLOS (non-line- of-sight) conditions, presented. The relative position between the vehicle and the endangered road users is carried out in the vehicle and is based on a radio frequency system. In addition, a LiDAR ultrasonic system can be used as an additional sensor. In the claimed method, a future position of a road user within a traffic space is determined using a movement vector. The movement vector can be verified based on a movement of the road user.

The WO 2017/054965 A1 shows and describes a device and a method for characterizing objects. A sending and a receiving vehicle can share a traffic space. In a dynamic object container, data of dynamic objects can be transmitted, which the sending vehicle has recognized with the help of its sensors. These can be pedestrians in particular. In particular, the transmitting vehicle can use LiDAR techniques to capture the vehicle environment. The receiving vehicle can use the recorded data for object recognition.

After all, she teaches US 2020/0023842 A1 a device having a memory to store an observed trajectory of a pedestrian. The pedestrian may be outside of a vehicle's field of view. The pedestrian can be tracked using infrastructure such as cameras, radars and LiDARs deployed at a traffic intersection even when there is no direct line of sight between the vehicle and the pedestrian. The trajectory of the pedestrian can be predicted. Warnings of a possible impending collision with the pedestrian can then be provided to the vehicle. However, it does not use the vehicle's LiDAR system to monitor the movement of the pedestrian without a direct line of sight between vehicle and pedestrian, but a LiDAR system that is part of the environment of the vehicle's LiDAR system or the road infrastructure and a has a direct line of sight to the pedestrian.

Disclosure of Invention

According to the invention, a LiDAR system of the type mentioned at the outset is made available, the LiDAR system being set up to use the light beam to obtain information about the object that is located in an area of the environment to which there is no direct line of sight for the LiDAR system. system exists.

Advantages of the Invention

The LiDAR system therefore has the advantage of being able to obtain information about the object from the light beam of the LiDAR system, which scans the environment, even though the object is outside the direct line of sight of the LiDAR system. This means that the light beam does not have to hit the object directly. The LiDAR system is able to scan hidden areas of the environment without resorting to active external aids. Active external aids would be, in particular, a vehicle in the vicinity, which makes sensor data available, for example, or another LiDAR system that is stationarily mounted on a street through which the present LiDAR system is moving.

In urban areas, occlusions, i.e. areas to which there is no direct line of sight for the LiDAR system, e.g. B. by parked vehicles often available. The area where there is no direct line of sight for the LiDAR system can also be referred to as an occlusion. These occlusions, hidden by obstructions to the LiDAR system's view, such as walls or parked vehicles, potentially hide other road users, e.g. B. pedestrians or children. The risk of a collision with them is reduced in the prior art by careful driving (speed and distance) and other methods (AI predictions or C2X, i.e. Car to X: vehicle communication with other vehicles, infrastructure or other types of information sources). An insight into the concealment via sensors, as proposed here, means that one can choose an optimized driving style and at the same time maintain or improve safety against VRUs (vulnerable road users - vulnerable road users such as pedestrians, cyclists, children). Exact knowledge of the geometry of the environment or extensive measurement or training is not necessary with the proposed LiDAR system, in contrast to the prior art.

The present LiDAR system thus preferably allows a response to an object in the environment even before the LiDAR system has a direct line of sight to the object, based solely on the light beam emitted into the environment by the LiDAR system itself . It goes without saying that within the meaning of this application it is irrelevant whether or not there is actually an object in the covered area to be monitored. In any case, the LiDAR system is set up to obtain information about such an object if this object is located in the area of the environment to which there is no direct line of sight for the LiDAR system. One can only know whether an object is actually in an occlusion or not after operating the LiDAR system for indirectly monitoring the occlusion without an existing direct line of sight into the occlusion.

Embodiments of the LiDAR system provide that the LiDAR system is set up to obtain information about a movement of the object, even though the object is in the area of the environment to which there is no direct line of sight. While a vehicle driver or an automatic driving mode can usually control the movement of their own vehicle well, the movement of objects in the vehicle's environment poses a particular risk of accidents. Objects can be in an area that is not for the driver or the automatic vehicle are directly visible. It is therefore particularly advantageous to monitor the environment for possible movements in occlusions to which there is no direct line of sight and which can therefore lead to unexpected dangerous situations. In contrast to the prior art, the proposed LiDAR system is designed to detect movements in selectable areas of the environment in a robust and rapid manner in complex situations. In some embodiments, this can be based on conventional LiDAR systems, which use the object in the environment to gain information tion would have to be sampled directly. However, those embodiments are preferred in which the LiDAR system is set up to also scan the environment directly. The LiDAR system is therefore preferably set up to carry out both line-of-sight measurements (ie with a direct line of sight) and non-line-of-sight measurements (ie without a direct line of sight).

In some embodiments, the LiDAR system is set up to indirectly monitor the area of the environment to which there is no direct line of sight for a lateral movement of the object. Lateral movements in particular often lead to traffic accidents in the automotive sector. Advantageously, this LiDAR system does not display an image of the object, as is known from the prior art, but merely represents a motion detector that monitors an area that can be precisely defined for lateral movements. This makes the method practical and robust. In the present case, lateral preferably means essentially perpendicular to the trajectory on which the LiDAR system is moving, such as perpendicular to a roadway. A vehicle in which the LiDAR system is arranged can move longitudinally across the lane and the object can move laterally to the lane, i.e. in the same plane as the vehicle, such as towards or away from the lane. Lateral is therefore preferably meant in the sense of an approach from the side area of a road. A pedestrian approaches, for example, laterally to the path of the—preferably automatic—vehicle that can have the LiDAR system.

The LiDAR system may be configured to select one or more scatter points in the environment and to detect movement of the object within a specified radius around the one or more selected scatter points, each of the scatter points being located in an area of the environment to which there is a direct line of sight for the LiDAR system. The scattering point in the roadway is preferably the position on the first reflecting surface, which is struck by the light beam from the LiDAR and from which the light beam is also scattered, among other things, into the covered areas. In poor conditions, the reduced performance can be read directly from the measured value of the scatter point (from the primary reflection of the scatter point on the ground). This can be used as a kind of control value to estimate the performance. In this way, blindness and degradation can be detected. If necessary, the driving speed can be adjusted to the performance. The radius is preferably set to between 5 centimeters and 1 meter, in particular between 20 centimeters and 80 centimeters, preferably between 30 centimeters and 60 centimeters. A particularly preferred radius is 50 centimeters. It is preferred that a large part of the area defined by the radius lies in the area of the environment to which there is no direct line of sight of the LiDAR system. In this way, any light that bounces back from the object after the light beam has indirectly illuminated it via the scattering point can be captured by the LiDAR system, no matter where in the radius the light strikes. In some embodiments, the scatter points are also optimally set by an algorithm on the basis of other information, preferably information from the map or from other sensors (preferably video sensors), in order to cover hazardous areas in the environment.

Some embodiments provide that the LiDAR system is set up to indirectly monitor the area of the environment to which there is no direct line of sight by scanning the area to which there is a direct line of sight for the LiDAR system with overlapping detection zones. one of the scattering points being located in the centers of the detection zones. The detection zones can be circular zones. Circular zones are preferably circular detection zones in a top view. Scanning can mean making multiple measurements adjacent to each other on surfaces in the environment to obtain the information about occluded areas from multiple perspectives or in different areas, particularly in a row and column pattern. If one carries out repeated measurements at different scattering points next to each other, then preferably circular zones result, for which one can make speed statements about hidden objects. In this way you can "scan" the edges of the open space and transfer them to a map with other outdoor information. If the detection zones are chosen to overlap, the intersection points of the circles can further improve the spatial resolution. A particularly advantageous measurement grid is thus generated.

In some embodiments, the LiDAR system is configured to select one or more of the scattering points at a distance of approximately 20 meters from the LiDAR system. This distance is preferably greater than 12 meters and less than 30 meters. In particular, the distance can be between 15 and 25 meters, preferably exactly 20 meters. Such distances are particularly advantageous for monitoring the movement of pedestrians, because when movement is detected, taking into account normal pedestrian speeds, there is still a sufficiently long reaction time at the distances mentioned to take measures to avoid a collision. If the distance is shorter than 12 meters in particular, it may not be possible depending on the driving situation enough time to take appropriate action to avoid a collision. In addition, for the distances mentioned, the LiDAR system does not require a large range and the power of the light beam can therefore remain relatively low. Due to the low optical transmission power adapted to the small distance, there are advantages, for example, in terms of eye safety, cooling of the signal source and also the lower power consumption of the system.

Some embodiments provide that the LiDAR system is set up to select one or more of the scattering points based on one or more selection criteria. If scatter points are selected and not left to chance, the LiDAR system can be used to select objects in the environment for the placement of the scatter points that have particularly good scatter properties. An active scatter point search for a comparatively good scatter point in a defined area that provides lateral scatter is advantageous. Reflections on puddles, for example, should be avoided. Clues as to which scattering points are well suited (reflectance, scattering characteristics) can be obtained from other sensor information. A paper cup on the side of the road or a bright marker can be an excellent scattering point for some embodiments.

It is preferred that the LiDAR system is an FMCW LiDAR system and has a device to move the light beam in a space, particularly preferred to move the light beam in a vertical plane. In this way, the LiDAR system can scan a vertical plane with the light beam, which can have a large number of detection zones. The space is preferably a space in front of a vehicle in which the LiDAR system is integrated. The spatial resolution of the LiDAR system can thus be improved. This provides a high resolution LiDAR system although only a single light source needs to be provided. Miniaturized line scanners have already been developed in the past. In principle, such a device can also determine distance and speed values using a method other than FMCW, namely using laser Doppler interferometry. The present invention could represent an application. The invention can thus be used in embodiments in connection with laser Doppler interferometry. The beam and detector can be moved in a vertical plane by the mirror and provide the necessary anticipation depending on the speed. The FMCW LiDAR system (or another suitable modulation) emits light in the frequency range that is not visible to the eye, which allows speed and distance measurement due to its modulation. The measuring principle preferably uses reflections. The knowledge of the reflection point should be used and as much signal as possible should be returned.

The LiDAR system preferably includes a transmission unit and a reception unit. The transmission unit preferably includes the light source, in particular a laser, in order to emit the light beam from the LiDAR system into the environment. In embodiments, the laser is an infrared laser. A version that emits a bundled beam is advantageous. Ideally, the angle of the light beam of the LiDAR system can be adjusted using mirrors, facial expressions on the transmitter or electrical beamforming by controlling the phases of several transmitter elements. One or more fixed transmission elements are also conceivable. It is also possible to "scan" the environment or part of the environment due to the vehicle's own movement, even with non-moving sensors or mirrors. A combination of one or more fixed systems together with one or more adjustable systems is also possible.

The receiving unit preferably comprises a light sensor, such as a photodiode, in order to receive light from the environment, in particular the light beam after it has been reflected in the environment. The receiving unit preferably has a high sensitivity (aperture, diaphragm, optics...). Signal processing close to the sensor advantageously allows a selective amplification of areas that match a search window in the distance-speed space. The receiving unit is preferably set up to be insensitive to ambient light, in particular by means of an optical filter, electrical suppression of the non-modulated background light and processing.

The LiDAR system is preferably set up to carry out transit time measurements using the light beam in order in particular to obtain depth information about the environment. The LiDAR system can be set up to calculate a distance between the LiDAR system and the object, preferably when the object is in an area of the environment to which there is a direct line of sight for the LiDAR system.

The LiDAR system may include a microprocessor and memory coupled thereto storing a program having instructions for performing a method of operating the LiDAR system using the microprocessor.

One aspect of the invention can be a vehicle that has a LiDAR system of the type described includes. The vehicle can be a motor vehicle, in particular a road-bound motor vehicle, for example a passenger car or a truck or a two-wheeler. In addition to the typical positions of sensors on the vehicle such as the windshield, radiator grille or impact protection, the LiDAR system can preferably be integrated in a headlight of the vehicle. In contrast to many sensors, the LiDAR system can be installed unobtrusively in the right and/or left headlight. With clever integration, a large aperture can also be implemented here without disturbing the design. In this way, the headlight has a second function as the housing for the LiDAR system, without impairing the actual function of the headlight.

1. 1. Streuung auf dem Streupunkt auf der Straße, insbesondere einem Gegenstand auf oder neben der Fahrbahn,
2. 2. Streuung am Objekt, wie etwa einem Fußgänger, und
3. 3. Streuung wieder in der Nähe des Streupunktes oder an einer spiegelnden Oberfläche.

The challenge with the present invention lies in the triple diffuse scattering, as this means that only very little useful signal returns to the receiver in the receiving unit of the LiDAR system:

1. 1. scattering on the scattering point on the road, in particular an object on or next to the roadway,
2. 2. Scattering at the object, such as a pedestrian, and
3. 3. Scattering again near the scattering point or on a specular surface.

• A_ped = 0.5^2 % area pedestrian [m*m]
• rho = 0.2 % reflectance asphalt [1]
• rho_ped = 0.4 % reflectance pedestrian [1]
• d_ped = 1; % distance pedestrain from SP [m]
• d_SP = [1:1:20]; % distance SP to Detector [m]
• A_SP3 = 1; % area 2. asphalt reflection [m*m]
• A_REC = 0.2*0.2 % area detector [m*m]
• % Simplified ‚Back-of-the-envelope calculation‘
• H1 = 1
• H2 = H1*rho
• H3 = H2*A_ped/(d_ped"2 * 2 *pi) % 2*pi - radiation in half-space only % A_sphere = 4*pi*R^2
• H4 = H3*rho_ped
• H5 = H4*A SP3/(d_ped^2 * 2 *pi)
• H6 = H5*rho
• H7 = H6*A_REC./(d_SP.^2 * 2 *pi) ; ratio = H7./H1

That limits the range. However, pedestrian protection does not require a large range (e.g. 20 meters in particular). From this rough reflection it follows that it is important to transmit a collimated beam, focusing sensitivity on the small effects in the higher frequency (extra path and velocity) to contrast with the strong primary reflection. The LiDAR system receiver should have a large aperture to capture these faint components. Attenuation to 10^-9 appears enormous, but a 100-meter, smaller-aperture, 'normal' LiDAR system designed to detect a small, non-traversable object on a roadway, such as a rock, experiences a similar attenuation. The attenuation factor is shown in the following rough calculation (roughly simplified approach of uniform radiation in the half-space):

• A_ped = 0.5^2 % area pedestrian [m*m]
• rho = 0.2% reflectance asphalt [1]
• rho_ped = 0.4 % reflectance pedestrian [1]
• d_ped = 1; % distance pedestrain from SP [m]
• d_SP = [1:1:20]; % distance SP to detector [m]
• A_SP3 = 1; % area 2. asphalt reflection [m*m]
• A_REC = 0.2*0.2 % area detector [m*m]
• % Simplified 'Back-of-the-envelope calculation'
• H1 = 1
• H2 = H1*rho
• H3 = H2*A_ped/(d_ped"2 * 2 *pi) % 2*pi - radiation in half-space only % A_sphere = 4*pi*R^2
• H4 = H3*rho_ped
• H5 = H4*A SP3/(d_ped^2 * 2 *pi)
• H6 = H5*rho
• H7 = H6*A_REC./(d_SP.^2 * 2 *pi) ; ratio = H7./H1

Three possible application examples for the proposed LiDAR system follow.

Example 1 - Pedestrians in city traffic - Pedestrians approach the roadway from a concealed position

wobei a_max die maximale Bremsverzögerung ist, t_lat die Latenzzeit bis zur Erkennung und Bremsreaktion auf den Fußgänger ist und v_ped die maximale Geschwindigkeit des Fußgängers ist.With the new method, the pedestrian can be recognized as a moving object even before the line of sight is reached. A doubling of the lateral visibility allows, if one neglects the latency, a twice as high speed of the vehicle with physical avoidability of a collision. $$v_{\text{Max}}=a_{\text{Max}}\left({\mathrm{i.e}}_{\text{lateral}}^{\text{front}}/v_{\text{ped}}-t_{\text{lat}}\right)$$

where a _max is the maximum braking deceleration, t _lat is the latency to detect and brake response to the pedestrian, and v _ped is the maximum speed of the pedestrian.

Example 2 - leaving a parking space in a confusing situation

The advantage is that you don't have to feel your way to the line of sight with the vehicle and therefore don't risk any body damage. The system advantageously also supplements a camera (preferably a near-range camera).

Example 3 - truck with trailer - checking the lateral space and the rear space for movement.

The advantage is that the sensor can be installed on a towing vehicle and can check the surroundings of a trailer. A sensor in the trailer is therefore not necessary.

The invention also provides the method for operating a LiDAR system of the type mentioned at the outset, which further according to the invention comprises the step of obtaining information about an object located in the environment using the light beam of the LiDAR system, wherein the object is located in an area of the environment to which there is no direct line of sight for the LiDAR system.

The method therefore has the advantage of being able to obtain information about the object from the light beam of the LiDAR system, which scans the environment, even though the object is outside the direct line of sight of the LiDAR system. This means that the light beam does not have to hit the object directly. The LiDAR system is able to scan hidden areas of the environment without resorting to active external aids. Active external aids would be, in particular, a vehicle in the vicinity, which makes sensor data available, for example, or another LiDAR system that is stationarily mounted on a street through which the present LiDAR system is moving.

In urban areas, occlusions, i.e. areas to which there is no direct line of sight for the LiDAR system, e.g. B. by parked vehicles often available. The area where there is no direct line of sight for the LiDAR system can also be referred to as an occlusion. These occlusions, hidden by obstructions to the LiDAR system's view, such as walls or parked vehicles, potentially hide other road users, e.g. B. pedestrians or children. The risk of a collision with these is reduced in the prior art through careful driving (speed and distance) and other methods (AI predictions or C2X). An insight into the concealment via sensors, as proposed here, means that one can choose an optimized driving style and at the same time maintain or improve safety against VRUs (vulnerable road users - vulnerable road users such as pedestrians, cyclists, children). Exact knowledge of the geometry of the surroundings or extensive measurement or training is not necessary with the proposed method, in contrast to the prior art.

It is preferred that the method comprises the step of setting a search speed and a search distance in the environment for the light beam by the LiDAR system in order to obtain the information about the object. In this way, operation can be optimized in such a way that a meaningful measurement signal is obtained.

A particular embodiment of the method is described in detail below. The method can be carried out in particular by an FMCW LiDAR system. The measuring principle used here makes it possible to detect a movement that is covered in the sense of the direct line of sight in a well-defined, i.e. fixed, radius around a scattering point.

• V₁ sei V_ego*cos α; α sei der Neigungswinkel des Lichtstrahls relativ zur Horizontalen.

In a first step, the method can initially include the emission of a measurement beam, ie the light beam, and the determination of the distance d _SP of the scattering point SP. Then, in a second step, the search speed and the search distance can be determined. Since the vehicle moves itself, the vehicle's own speed V _ego must be taken into account:

• Let V ₁ be V _ego *cos α; α is the angle of inclination of the light beam relative to the horizontal.

Velocity interval V _range = [V ₁ + V _{Object_min} ; V ₁ + V _{_object_max} ], where V _{_object_max} ~ 7km/h for pedestrians and V _{_Object_min} ~ 2km/h. $${\text{Distance interval D}}_{\text{range}}=\left[{\text{i.e}}_{\text{SP}}+\mathrm{\delta };{\text{i.e}}_{\text{SP}}+\text{R}\right]$$

R ~ 2 meters to see one vehicle width into the cover; δ should be chosen as large as possible to separate the indirect signal from the very strong direct reflection.

Then, in a third step, the sensitivity thresholds can be adjusted to the search speed V _range and search distance D _range in the signal processing close to the sensor. Then, in a fourth step, the signals (locations) are preferably determined in the distance-velocity space ( _Varge and _Drange ) using FMCW methodology. A robust evaluation of the locations then preferably follows in a fifth step, in particular using a voting algorithm based on a number of measurements or a threshold value comparison of the quality criteria. Subsequently, in a sixth step, further processing can be carried out, if necessary, preferably by sensor fusion with outdoor sensors or "region of interest" with increased sensitivity for outdoor sensors (e.g. partially covered pedestrians for video). Finally, in a seventh step, one or more measures from the group consisting of reducing the speed of a vehicle, braking the vehicle and preparing for the intervention by applying the brake shoes of the vehicle can be triggered. In the active search, steps 1-4 can be performed repeatedly until corresponding signals are detected in the V _rarge , D _range . Provision is preferably made for the steps to be carried out in the order mentioned.

The method can be structured in instructions forming a computer program. The computer program may be stored in memory to instruct an associated microprocessor configured to interpret the program to perform the method of operating the LiDAR system.

Further refinements of the method and their advantages result mutatis mutandis from the above description of the LiDAR system, to which reference is made here in full in order to avoid repetition. In embodiments, the described LiDAR system can be set up to carry out the described method for operating a LiDAR system.

Advantageous developments of the invention are specified in the dependent claims and described in the description.

drawings

• 1 ein LiDAR-System in einer ersten Ausführungsform, das in einem Fahrzeug integriert ist;
• 2 ein Ablaufdiagramm einer ersten Ausführungsform eines Verfahrens zum Betreiben des LiDAR-Systems, und
• 3 ein Ablaufdiagramm einer zweiten Ausführungsform des Verfahrens zum Betreiben des LiDAR-Systems.

Exemplary embodiments of the invention are explained in more detail with reference to the drawings and the following description. Show it:

• 1 a LiDAR system in a first embodiment integrated in a vehicle;
• 2 a flowchart of a first embodiment of a method for operating the LiDAR system, and
• 3 a flowchart of a second embodiment of the method for operating the LiDAR system.

Embodiments of the invention

In the 1 a LiDAR system 1 is shown in a first embodiment, which is integrated into a vehicle 2 . The vehicle 2 is here, for example, a fully autonomous electric vehicle for passenger transport. The LiDAR system 1 is installed in a suitable area 3 of the vehicle 2, which represents a housing for the LiDAR system 1, here purely by way of example in a headlight 3. An arrow F indicates a direction of travel of the vehicle 2.

The LiDAR system 1 is set up to scan an environment of the LiDAR system 1 with a light beam 4 in order to obtain depth information about the environment. The LiDAR system 1 is also set up to obtain information about an object 5 that is in the environment from the light beam 3 . As shown, the object 5 is in an area of the environment to which there is no direct line of sight for the LiDAR system 1 . The object 5 is here, for example, a person 5, shown schematically, who is hidden behind a visual obstacle 6, shown schematically here, a wall 6, for example. The wall 6 thus interrupts the direct line of sight between the LiDAR system 1 and the person 5. A light source (not shown) of the LiDAR system 1, which is set up to scan the environment with the light beam 4, is therefore not in in the situation shown able to direct the light beam 4 directly, i.e. in a straight line without deflection, onto the person 5 in order to scan him and thus detect his presence. Therefore, according to the prior art, an occupant of the vehicle 2 would either have to resort to external systems in order to recognize the person 5, or approach the wall 6 with extreme caution, since he cannot immediately know whether a person 5 is behind the wall 6 is and could possibly run from there into a route 7 of the vehicle 2 .

However, the present LiDAR system 1 is set up to use the light beam 4 to obtain information about the object 5 that is located in an area of the environment to which the LiDAR system 1 has no direct line of sight. More precisely, the LiDAR system 1 is set up to obtain information about a movement of the object 5, although the object 5 is in the area of the environment to which there is no direct line of sight for the LiDAR system 1. Specifically, the LiDAR system 1 is set up to indirectly monitor the area of the environment to which there is no direct line of sight for a lateral movement of the object 5 . Thus, despite the lack of a direct line of sight, the occupant of the vehicle 2 can be warned if the person 5 moves behind the wall 6 in the direction of the route 7 of the vehicle 2 .

To achieve this, the LiDAR system 1 is set up to select one or more scattering points 8 in the environment and to detect the movement of the object 5 in a specified radius around the one or more selected scattering points 8, with each of the Scatter points 8 is in an area of the environment to which there is a direct line of sight for the LiDAR system 1. In this case, the LiDAR system 1 is set up to select the scatter point or points 8 based on one or more selection criteria. In the present example, the LiDAR system 1 has selected an object 9, namely a coffee mug 9, which lies on the roadway 7, for the placement of the scattering point 8 because the coffee mug 9 has a suitable reflectance. Furthermore, the LiDAR system 1 is set up to select one or more of the scatter points 8 at a distance of approximately 20 meters from the LiDAR system 1 . Experience has shown that many accident risks due to moving objects 5 such as other vehicles, people or animals in the environment arise at this distance from the vehicle 2 and this distance still allows a suitable reaction to a detected hidden movement. The coffee mug 9 then scatters the light beam 3 behind the wall 6 onto the person 5, from where the light beam 3 falls back onto the roadway 7, in a specified radius around the scattering point 8, and from there it is reflected back to the LiDAR system 1. The specified radius is determined by the optics and the nature of the receiver and is around 30 cm here.

More specifically, the LiDAR system 1 is configured to indirectly monitor the area of the environment to which there is no direct line of sight by scanning the area to which there is a direct line of sight for the LiDAR system with overlapping detection zones, wherein one of the scatter points such as scatter point 8 is located in the centers of the detection zones. Further scatter points at locations different from the scatter point 8 are thus detected by the LiDAR system 1 correspondingly with overlapping detection zones, here by way of example circular zones, by means of the light beam 4 according to the same principle, i.e. quasi scanned. The resolution is thus increased by the scanning, because the various measurements can be combined, e.g. by triangulation, and thus improved information of the covered area behind the wall 6 by means of the light beam 4 results. The LiDAR system 1 is an FMCW LiDAR system 1, which has a device to move the light beam 4 in a vertical plane, here in particular vertical to a radiation direction of the LiDAR system 1. So the scanning of stray points 8 on or can be reliably reached next to the roadway 7 for indirect scanning of covered areas of the environment.

In 2 a method for operating a LiDAR system 1 is schematically illustrated. As a first step S21, the method includes scanning the environment of the LiDAR system 1 using a light beam 4 of the LiDAR system 1. As a second step S22, the method includes obtaining information about an object 5 that is located in the environment , by means of the light beam 4 of the LiDAR system 1, the object 5 being located in an area of the environment to which there is no direct line of sight for the LiDAR system 1. The LiDAR system 1 off 1 is set up to carry out this method by means of a microprocessor which is not shown. Another possible step of the method, which is not illustrated here, is the LiDAR system 1 determining a search speed and a search distance in the environment for the light beam 4 in order to obtain the information about the object 5 . For example, the method can provide for finding a suitable stray point 8 at a search distance of 20 meters, for example with a search speed of 3 searches per second. This means, for example, that the environment is scanned three times per second at a distance of 20 meters from the LiDAR system 1 in order to find a suitable point on the road or an object 9 in the roadway 7 on which the scatter point 8 can be placed. If no suitable object that reflects sufficiently is found within a definable number of searches, the distance and/or the speed of the search can be adjusted.

In 3 another method is illustrated, which is based in principle on the method 2 based and refined. Also the procedure 3 can through the LiDAR system 1 1 to be executed. In a first step S31, the method provides for a measurement beam to be emitted, namely the light beam 3, and for the distance of the scattering point 8 to be determined. This means the distance between LiDAR system 1 and scatter point 8. In a second, subsequent step S32, the method provides for setting the search speed and the search distance. In a third, subsequent step S33, the method provides for an adaptation of the sensitivity thresholds to the search speed and the search distance in the signal processing close to the sensor. In a fourth, subsequent step S34, the signals in the distance-speed space are determined using FMCW methodology. In a fifth, subsequent step S35, the signals are evaluated. Depending on the application, this can be done using a voting algorithm or a threshold value comparison of the quality criteria. Further processing can take place in a sixth, subsequent step S36. Finally, in a seventh step S37, which concludes the exemplary method here, one or more measures can be triggered on the vehicle 2, such as speed reduction, braking or preparation for the intervention by applying brake shoes of the vehicle 2. Steps S31 to S34 can be carried out repeatedly in a variation of the method until corresponding signals are detected in V _rarge , D _range .

The invention thus provides a LiDAR sensor 1 and a method for predictive motion detection in a covered area that can be precisely defined, in particular for automatic driving in urban areas. The non-line-of-sight sensing sensor of the LiDAR system 1 is attached to a vehicle 2 (e.g. as in 1 on a vehicle front). In embodiments, the LiDAR system 1 is an active optical sensor that measures distance and speed, which preferably has at least one bundled measuring beam, i.e. light beam 4, with a measuring width or range of, for example, 20 meters (~ pedestrian monitoring area 5 in of the city) which allows non-line-of-sight sensing.

An additional meter of lateral distance can significantly improve emergency braking in difficult urban scenarios. A pedestrian 5 running out of a concealment at 5 km/h, in particular behind a wall 6, could be reacted to 700 ms earlier, for example, with 1 meter more lateral visibility. In reality, there is often concealment, which is why it can be found in accidents and is therefore also checked and evaluated in this way in Euro-NCAP scenarios.

This LiDAR system 1 can be used in particular as an additional independent sensor to monitor the edge of the roadway 7 for lateral movements in covered areas of the environment. Methods for LiDAR measurements at large distances (amplification, signal processing with poor SNR) can be transferred to the new non-line-of-sight sensing, i.e. scanning without a direct line of sight, over short distances. This opens up a completely new sensing option in embodiments, especially for urban traffic, in particular for highly automated vehicles 2 (shuttle, robo-taxi ...) but also for L2 vehicles 2. The approval discussion L3, L4 can be System 1 are supported.

Although the invention has been illustrated and described in more detail by means of preferred exemplary embodiments, the invention is not restricted by the disclosed examples and other variations can be derived therefrom by a person skilled in the art without departing from the protective scope of the invention.

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• DE 102014207896 A1 [0005]
• DE 112019000057 T5 [0006]
• DE 102014219165 A1 [0007]
• WO 2017/054965 A1 [0008]
• US 2020/0023842 A1 [0009]

Claims (10)

LiDAR system (1) that is set up to scan an environment of the LiDAR system (1) with a light beam (4) in order to obtain depth information about the environment, the LiDAR system (1) being further set up to using the light beam (4) to obtain information about an object (5) located in the environment, characterized in that the LiDAR system (1) is set up to use the light beam (4) to obtain information about the object ( 5) that is in an area of the environment where there is no direct line of sight for the LiDAR system (1). LiDAR system (1) after claim 1 , wherein the LiDAR system (1) is set up to obtain information about a movement of the object (5), although the object (5) is in the area of the environment to which there is no direct line of sight for the LiDAR system ( 1) exists. LiDAR system (1) after claim 1 or 2 , wherein the LiDAR system (1) is set up to indirectly monitor the area of the environment to which there is no direct line of sight for a lateral movement of the object (5). LiDAR system (1) according to one of claims 2 or 3 , wherein the LiDAR system (1) is set up to select one or more scattering points (8) in the environment and to detect the movement of the object (5) in a specified radius around the one or more selected scattering points (8). , wherein each of the scattering points (8) is located in an area of the environment to which there is a direct line of sight for the LiDAR system (1). LiDAR system (1) after claim 4 , wherein the LiDAR system (1) is set up to indirectly scan the area of the environment to which there is no direct line of sight by scanning the area to which there is a direct line of sight for the LiDAR system (1) with overlapping detection zones monitor, with one of the scattering points (8) being located in the centers of the detection zones. LiDAR system (1) according to one of Claims 4 or 5 , wherein the LiDAR system (1) is set up to select one or more of the scattering points (8) at a distance of approximately 20 meters from the LiDAR system (1). LiDAR system (1) according to one of Claims 4 until 6 , wherein the LiDAR system (1) is set up to select one or more of the scattering points (8) based on one or more selection criteria. LiDAR system (1) according to any one of the preceding claims, wherein the LiDAR system (1) is an FMCW LiDAR system (1) having means for moving the light beam (4) in a space. Method for operating a LiDAR system (1), comprising the step - Scanning (S21) an environment of the LiDAR system (1) by means of a light beam (4) of the LiDAR system (1), the method further comprising the step: - Obtaining (S22) information about an object (5) located in the environment by means of the light beam (4) of the LiDAR system (1), the object (5) being located in an area of the environment where there is no direct line of sight for the LiDAR system (1). procedure after claim 9 , comprising the step of: - determining a search speed and a search distance in the environment for the light beam (4) by the LiDAR system (1) in order to obtain the information about the object (5).

Images