Method for locating a perception sensor on a vehicle

Abstract

The invention provides a method of locating a perception sensor on a vehicle. A three-dimensional overlay of the perceptual sensors is determined, and vehicle-specific geometric features are received. The coverage area of the perceptual sensor in the vicinity of the vehicle is estimated by combining vehicle-specific geometric features and a three-dimensional coverage map of the perceptual sensor and by changing the spatial position of the perceptual sensor until the coverage area in the vicinity of the vehicle is optimized.

Description

Technical Field

[0001] The present disclosure relates to a computer-implemented method of positioning perception sensors on a vehicle. Background Art

[0002] In order to perform a specific maneuver of the vehicle, the vehicle's external environment must be monitored in an appropriate manner as a whole. To monitor the vehicle's external environment, perception sensors such as radar systems, lidar systems or cameras are used. For example, if an autonomous parking application is to be performed, it is necessary to provide physical evidence of the feasibility of such an application, which can be achieved by a radar system covering both close and long distances relative to the vehicle.

[0003] For known radar systems, a two-dimensional coverage map is usually generated, in which certain tolerances regarding the spatial installation position of the radar sensor are taken into account, such as tolerances of the roll angle, pitch angle and yaw angle of the radar sensor. Such a two-dimensional coverage map properly covers large distances relative to the vehicle and can be suitable for applications that only require such large distances relative to the vehicle.

[0004] However, known two-dimensional coverage maps for radar sensors installed in vehicles do not provide information for very short distances that may be required for low speed and parking applications. Furthermore, known two-dimensional coverage maps do not take into account vehicle-specific design features. Due to the specific mounting of the radar sensor on the vehicle, vehicle components may cause shielding effects of the radar sensor that cannot be identified by the known two-dimensional coverage maps. Furthermore, the effect of the considered specific mounting height of the radar sensor on the spatial coverage of the radar sensor is usually not included in the two-dimensional coverage map. Therefore, by using known coverage maps, a proper radar integration that makes the radar system suitable for supporting autonomous parking applications cannot usually be performed.

[0005] Therefore, a method of integrating a perception sensor in a vehicle is needed so that the perception sensor can support low-speed applications and autonomous parking applications of the vehicle. Summary of the invention

[0006] In one aspect, the present invention relates to a computer-implemented method for positioning a perception sensor on a vehicle. According to the method, a three-dimensional coverage map of the perception sensor is determined and a vehicle-specific geometry is received. A coverage area of ​​the perception sensor near the vehicle is estimated by combining the vehicle-specific geometry and the three-dimensional coverage map of the perception sensor and by changing the spatial position of the perception sensor until the coverage area near the vehicle is optimized.

[0007] The three-dimensional coverage map may first be determined independently of a specific vehicle or vehicle line, and then the three-dimensional coverage map may be scaled to vehicle coordinates. If the perception sensor includes a radar sensor, for example, when determining the three-dimensional coverage map, a target coverage range of a radar cross section (RCS) of -15 dBsm (decibels per square meter) may be considered for the external environment of the sensor.

[0008] In addition to or as an alternative to radar sensors, the perception sensors may also include lidar sensors or camera systems. Thus, when determining the three-dimensional coverage map, the instrument field of view of the perception sensors may also be considered.

[0009] Vehicle-specific geometric features may be provided, for example, by CAD data relating to the outer contour of the vehicle. CAD data may also be taken into account when scaling the three-dimensional overlay relative to a specific vehicle or vehicle shape.

[0010] Optimizing the coverage area may include maximizing the coverage of the perception sensors in a desired predetermined area or volume near the vehicle, i.e., such that the uncovered so-called blind volume may be minimized, while the coverage of the perception sensors may be reduced in other areas or volumes outside the desired predetermined area or volume. The desired predetermined area or volume may depend on the specific application involving the perception sensors, such as a specific low-speed and / or parking application.

[0011] One advantage of this method is that it is possible to determine a coverage map in the vicinity of the vehicle that can represent a specific vehicle shape. Therefore, the spatial position of the perception sensor on the vehicle can be optimized by the coverage map in the vicinity of the vehicle so that low-speed scenarios and parking applications can be properly supported. The spatial position to be changed can include the position (i.e., the longitudinal and lateral coordinates of the perception sensor relative to the vehicle coordinate system and its installation height relative to the ground plane), and the angular orientation including the yaw angle, pitch angle and roll angle.

[0012] In the early stages of vehicle design, the method also enables the planning of parking configurations based on appropriate spatial locations of perception sensors. Conversely, the method can identify points or areas with weak or low coverage for specific vehicle shapes and the effectiveness of compensating such areas or regions with low coverage, for example by additional perception sensors, can be evaluated. For example, if the overall perception system includes multiple radar sensors, the integration scenario of these radar sensors can be optimized by the unified coverage map provided by the method.

[0013] According to an embodiment, changing the spatial position of the perception sensor may include changing the positioning height of the perception sensor relative to the ground plane and changing the yaw, pitch and roll angles of the perception sensor relative to the vehicle. The yaw, pitch and roll angles may be defined relative to a vehicle coordinate system having an origin, for example, at the center of gravity of the vehicle and having one axis aligned with the longitudinal axis of the vehicle.

[0014] For this embodiment, the positioning or mounting height of the perception sensor can only be changed when the longitudinal and lateral coordinates of the perception sensor relative to the vehicle can be fixed. Alternatively, the longitudinal and lateral coordinates of the perception sensor relative to the vehicle can also be changed to optimize the coverage area.

[0015] Due to the variation of positioning height and orientation (i.e., the angle of the perception sensor), in order to optimize the coverage area near the vehicle, the method can result in an optimized spatial position of the perception sensor in terms of its positioning height and yaw, pitch, and roll angles. For example, the coverage area can be maximized near the vehicle close to the optimized spatial position of the perception sensor, so that the perception sensor located at the optimized spatial position can properly support low-speed and parking applications.

[0016] According to another embodiment, the perception sensor may include a radar sensor, and determining the three-dimensional coverage map may include applying a link budget equation and a predetermined radar cross section. Due to this framework for determining the three-dimensional coverage map, a lower computational effort may be required to perform the method.

[0017] The optimal coverage area may be determined by using a predetermined base frequency of the radar sensor in order to determine an optimal spatial position of the radar sensor. Furthermore, a corresponding three-dimensional coverage map may be determined for each of a predetermined number of frequencies of the radar sensor that differ from the base frequency. By applying the corresponding three-dimensional coverage map, a corresponding coverage area of ​​the radar sensor may be estimated in the vicinity of the vehicle for each of the predetermined number of frequencies of the optimal spatial position of the radar sensor.

[0018] For example, the basic frequency may be 77 GHz, and other frequencies different from the basic frequency may include a predetermined number of frequencies in a range between, for example, 76 GHz and 81 GHz. Determining the corresponding additional coverage areas of the perception sensor at different frequencies for the optimized spatial position of the radar sensor may allow an analysis of whether a specific target near the vehicle can be detected at all frequencies of the frequency range defining the predetermined number of frequencies. This may be relevant for radar sensors that transmit radar waves at a fairly wide range of frequencies. In addition, it may be determined whether such a radar sensor allows detection of a specific target near the vehicle only within a specific frequency range that is smaller than the frequency range of the transmitting radar sensor.

[0019] The vehicle-specific geometry may include a CAD model of the vehicle (from computer-aided design CAD). Thus, the vehicle-specific geometry may be provided in a standardized format based on CAD, which may reduce the effort of receiving the vehicle-specific geometry for the method.

[0020] By taking into account the geometry provided by the CAD model of the vehicle, shadow areas in a predetermined area of ​​interest that are outside the coverage area may be determined. Shadow areas may be defined as areas that are, for example, located at ground level and are not covered or illuminated by the perception sensor due to being obstructed from the line of sight relative to the perception sensor by a part of the vehicle. Thus, the method may be able to assess whether certain areas within a predetermined area of ​​interest, for example relevant for low-speed and parking applications, are covered or illuminated by the perception sensor. This may support the design phase of the vehicle, for example at an early stage, where specific components of the vehicle are still flexible with regard to their position and / or design, and allow modifications to the position and / or design such that shadow areas may be reduced or even avoided.

[0021] Furthermore, at least one blind spot volume located outside the coverage area may be determined relative to the CAD model of the vehicle. In addition to shadow areas that may be caused by parts or components of the vehicle and may cause blind spot volumes, at least one further blind spot volume may be determined that may be caused by the spatial position of the perception sensor itself. Thus, the spatial position of the perception sensor may be substantially changed such that at least one blind spot volume may be appropriately reduced or even completely avoided.

[0022] In another aspect, the present disclosure relates to a computer system configured to perform several or all of the steps of the computer-implemented method described herein.

[0023] The computer system may include a processing unit, at least one memory unit, and at least one non-transitory data storage device. The non-transitory data storage device and / or the memory unit may include a computer program for instructing the computer to perform some or all steps or aspects of the computer-implemented method described herein.

[0024] As used herein, terms such as processing unit and module may refer to an application specific integrated circuit (ASIC), an electronic circuit, a combinational logic circuit, a field programmable gate array (FPGA), a processor (shared, dedicated, or group) that executes code, other suitable components that provide the described functionality, or a combination of some or all of the above, such as in a system on a chip, a portion thereof, or including them. A processing unit may include a memory (shared, dedicated, or group) that stores code executed by the processor.

[0025] In another aspect, the invention relates to a vehicle comprising a perception system with at least one perception sensor, wherein the spatial position of the perception sensor relative to the vehicle has been determined by performing a method as described above.

[0026] In this way, the perception system of the vehicle relies on perception sensors having spatial positions optimized by performing the above-described method steps. Therefore, the benefits, advantages and disclosure of the method are also valid for the vehicle according to the present disclosure.

[0027] According to one embodiment, the spatial position or orientation of at least one perception sensor has been determined such that at least one blind spot volume within a desired coverage area or region of interest of the perception sensor can be optimized at the vehicle. In other words, the blind spot volume at the vehicle can be optimized or minimized with respect to a specific application like parking or low speed applications in a desired predetermined area near the vehicle. In this way, it can be ensured that the vehicle is able to perform the specific application.

[0028] The perception system may include at least two perception sensors that may have overlapping three-dimensional coverage areas. Within such overlapping three-dimensional coverage areas, a particular target may be visible to more than one perception sensor. In this way, the confidence in detecting a particular target may be increased.

[0029] According to another embodiment, the perception system of the vehicle may include a radar system having at least one radar sensor. The radar system may include at least one front radar sensor, at least one radar sensor located on the side of the vehicle, and at least one rear radar sensor. For such a vehicle having at least three radar sensors, the respective spatial position of each radar sensor may be changed according to the method described above, so that the shadow area and / or any blind spot volume around the vehicle may be optimized as a whole for a particular group of radar sensors. Thus, by optimizing the spatial positions of all radar sensors together, the vehicle may be optimized for specific applications like low speed and parking applications.

[0030] Additionally or alternatively, the perception system may include a lidar system having at least one lidar sensor. For such a lidar sensor, the method according to the present disclosure may also be performed in a similar manner as described above for the radar sensor. If a lidar sensor is installed in addition to the radar sensor, the reliability of the perception system may be improved when detecting specific targets near the vehicle.

[0031] In another aspect, the present disclosure relates to a non-transitory computer-readable medium comprising instructions for performing some or all steps or aspects of the computer-implemented methods described herein. The computer-readable medium may be configured as: an optical medium, such as a compact disc (CD) or a digital versatile disc (DVD); a magnetic medium, such as a hard disk drive (HDD); a solid-state drive (SSD); a read-only memory (ROM); a flash memory; etc. In addition, the computer-readable medium may be configured as a data storage device accessible via a data connection such as an Internet connection. The computer-readable medium may be, for example, an online data repository or cloud storage.

[0032] The present disclosure also relates to a computer program for instructing a computer to perform several or all steps or aspects of the computer-implemented method described herein. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Exemplary implementations and functionality of the present disclosure are described herein in conjunction with the following drawings, which schematically illustrate:

[0034] Figure 1 (a) and (b) show the coverage map and coverage map scaling of the radar sensor;

[0035] Figure 2 (a) and (b) are diagrams for determining the coverage area of ​​the radar sensor near the vehicle;

[0036] Figure 3 (a), (b), (c) and (d) are three-dimensional illustrations of optimization of the blind spot volume at a particular vehicle, showing non-optimized and optimized coverage areas at the rear axle;

[0037] Figure 4 (a) and (b) are detailed two-dimensional plan views of the respective coverage areas and blind spot volumes of multiple radar sensors mounted on a vehicle, showing the non-optimized and optimized coverage areas at the rear axle;

[0038] Figure 5 is a flow chart illustrating a method for positioning a perception sensor on a vehicle according to various embodiments;

[0039] Figure 6 Systems according to various embodiments are shown; and

[0040] Figure 7 is a computer system having a plurality of computer hardware components configured to perform the steps of the computer-implemented method as described herein.

[0041] Reference numerals list

[0042] 100 Overlay

[0043] 110 Radar Sensor

[0044] 112 Radar sensor field of view

[0045] 114, 116 Field of view boundaries

[0046] 120 Vehicles

[0047] 150 Surface mesh of the overlay map

[0048] 160 Coordinate system

[0049] 170 Illuminated surface at ground level

[0050] 172 Profile of the illuminated surface

[0051] 200 CAD models of vehicles

[0052] 210 Ground Plane

[0053] 220 Coverage Area

[0054] 225 Projection Line

[0055] 230 Shadow Area

[0056] 240 Blind zone distance

[0057] 300 Total blind area volume

[0058] 310 Graphic height of blind area volume

[0059] 320 Specific blind spot volume at the vehicle's rear axle

[0060] 330 Specific blind spot volume at the front axle of the vehicle

[0061] 410-440 Main radar sensor coverage area

[0062] 450 Front gap-filling radar sensor coverage area

[0063] 460 Rear Gap Filler Radar Sensor Coverage

[0064] 472-478 Coverage area of ​​a radar sensor mounted on one side of a vehicle

[0065] 500 is a flow chart showing a method for locating a perception sensor on a vehicle

[0066] 502 Steps for determining a three-dimensional coverage map of a perception sensor

[0067] 504 Steps for receiving vehicle specific geometry

[0068] 506 By combining vehicle-specific geometric features and the 3D coverage map of the perception sensors and

[0069] Change the spatial position of the perception sensors until the coverage area near the vehicle is optimized.

[0070] Steps to estimate the coverage area of ​​the perception sensors near the vehicle

[0071] 600 Perception Sensor Positioning System

[0072] 602 Coverage Map Determination Circuit

[0073] 604 Vehicle Geometry Receiver Circuit

[0074] 606 Coverage Area Estimation Circuit

[0075] 608 Combinational Circuits

[0076] 610 Change Circuit

[0077] 611 Connections

[0078] 700 Computer Systems

[0079] 702 processor

[0080] 704 Memory

[0081] 706 Non-transitory data storage device

[0082] 708 connections DETAILED DESCRIPTION

[0083] Figure 1 (a) depicts a three-dimensional coverage map 100 determined for a radar sensor 110 as a first step of a method according to the present disclosure. The coverage map 100 is a three-dimensional spherical map that is estimated based on a link budget equation that takes into account a predetermined target radar cross section. For example, a target with a radar cross section (RCS) of -15 dBsm corresponding to, for example, a curb may be assumed. The three-dimensional coverage map 100 is estimated relative to a predetermined signal-to-noise ratio of the radar sensor and a minimum required signal-to-noise ratio that may be given by a swirling model.

[0084] The radar sensor 110 includes Figure 1 The predetermined instrument field of view 112 is indicated by the boundaries of lines 114 and 116 in (a). In addition, a vehicle 120 is shown to illustrate that the radar sensor 110 can be used as a corner radar sensor in the front area of ​​the vehicle 120.

[0085] As a next step according to the method of the present disclosure, Figure 1 The overlay 100 provided in (a) is generated to scale the three-dimensional CAD model. By using, for example, MATLAB, the three-dimensional overlay 100 is generated Figure 1 1 . The surface mesh 150 is represented by a plurality of crosses and is depicted relative to a coordinate system 160. Since the mounting height of the radar sensor 110 relative to the ground plane is known, an illuminated surface 170 at the ground plane can be determined, and the illuminated surface is shown in a boundary or contour 172 in the plane of z=0. Therefore, the illuminated surface 170 is covered by the radar waves emitted by the radar sensor 110. In addition, the mathematical representation of the illuminated surface 170 can be derived as a surface or polygon representing the contour 172 of the illuminated surface 170, for example, from MATLAB for further processing for other applications.

[0086] As a next step of the method according to the present disclosure, vehicle-specific geometric features are received in the form of a CAD model 200 of the vehicle 120. The CAD model 200 of the vehicle 120 is Figures 2 to 4 1 is shown as a line mesh including polygons representing different components on the exterior surface of the vehicle 120. Figure 2 In (a) of FIG. 1 , the rear of the vehicle 120 is shown. In the middle of the rear of the vehicle 120, the radar sensor 110 is installed at a specific installation height above the ground plane 210. Figure 2 As shown in (a), the coverage of the radar wave at -15 dBsm is depicted as a coverage area 220 represented by planes extending parallel to the ground plane 210 and having respective height distances of 10 cm from each other in a direction perpendicular to the ground plane 210.

[0087] The polygon is stretched out from the position of the radar sensor 110 until most of the outer part of the vehicle 120 is swept. The projection of the position of the radar sensor is stretched onto the road, that is, stretched to the ground plane 210. These projections are Figure 2 is shown as line 225 in (a).

[0088] As in Figure 2 (a) and in more detail in Figure 2 As can be seen in (b) of FIG. 1 , certain components of the vehicle 120 generate a shadow region 230 on the ground plane 210, which prevents the radar waves emitted by the radar sensor 110 from reaching the ground plane 210. This shielding effect of the vehicle components results in a blind distance on the ground 210 of approximately 260 mm behind the vehicle 120. In addition, the corresponding blind distances are shown for different target heights above the ground plane 210 (in steps of 10 cm). At a target height of 30 cm, the blind distance is reduced to approximately 141 mm.

[0089] As a further step of the method according to the present disclosure, the coverage area of ​​the radar sensor 110 in the vicinity of the vehicle 120 is estimated by combining the vehicle specific geometry (i.e., the CAD model 200 provided for the vehicle 120) and the three-dimensional coverage map 100 of the radar sensor 110. Figure 1 As shown in (b), a three-dimensional coverage map in the form of a surface grid 150 is provided. The coverage area of ​​the radar sensor 110 includes, for example, Figure 2 The three-dimensional volume is shown by parallel planes 220 in FIG. 2 , which are depicted as multiple layers at different heights relative to the ground plane 210. In addition, in Figure 4 4 also depicts coverage areas 410 to 478 of the different radar sensors installed on vehicle 120 as a plan view at a height of 30 cm relative to ground level 210 .

[0090] As another step of the method according to the present disclosure, the spatial position of radar sensor 110 is changed until the coverage area near vehicle 120 is optimized. In this example, the spatial position of radar sensor 110 includes the positioning height of radar sensor 110 relative to ground plane 210 and includes the spatial orientation of radar sensor 110 relative to vehicle 120 in terms of yaw, pitch, and roll angles. Figure 3 and Figure 4 Optimization of the coverage area near the vehicle 120 is shown in FIG.

[0091] exist Figure 3 In (a) and (b), the rear view and the front view of the vehicle 120 and the total blind volume 300 of the radar sensors installed at different positions on the vehicle 120 are schematically shown respectively. The total blind volume 300 is the volume around the vehicle 120 that is not illuminated by any radar sensor installed at the vehicle 120. In other words, the radar waves emitted by the radar sensor do not cover the blind volume 300. Figure 3 In (a) to (d) of FIG. 1 , the blind volume 300 is depicted at ground level 210 (see FIG. 2 ). Figure 2 ) up to a height of about 1m. Figure 3 In (d) the arrow 310 and the dashed line illustrate the height of about 1 m up to which the blind spot volume 300 is estimated. Figure 3 , vehicle 120 is represented by CAD model 200 .

[0092] For Figure 3 The total blind spot volume 300 shown in (a) and (b) of FIG. 1 is not changed in the following manner: relative to the optimized coverage area 220 for low speed and parking applications (see FIG. 1 ). Figure 2). Thus, the total blind spot volume 300 includes a substantial specific blind spot volume 320 proximate the rear axle of the vehicle 120. Thus, the perception system of the vehicle 100 including the radar sensor is blind around the rear wheels of the vehicle 120, which may prevent proper execution of low speed and parking applications.

[0093] However, for Figure 3 In the vehicle 120 shown in (c) and (d) of FIG. 1 , the spatial position of the radar sensors mounted on the vehicle 120 is varied in such a way that the coverage area near the rear axle is optimized for low speed and parking applications. Figure 3 As can be appreciated from (c) and (d) of FIG. 1 , at the rear axle of the vehicle 120, Figure 3 The specific blind spot volume 320 shown in (a) and (b) of the embodiment is almost completely removed. However, by optimizing the coverage area 220 of the radar sensor 110 with respect to low-speed and parking applications, the specific blind spot volume 330 at the front axle of the vehicle 120 is increased. However, this may not hinder the correct execution of these applications.

[0094] In detail, on the one hand Figure 3 The spatial position of the radar sensor for the vehicle 120 shown in (a) and (b) is different from that shown in (a) and (b) of FIG. Figure 3 The most important difference between the spatial positions of the radar sensors for vehicle 120 shown in (c) and (d) is the yaw angle of the so-called gap-filler radar mounted on the side of vehicle 120. Figure 3 For the vehicle 120 shown in (a) and (b), a yaw angle of 0° has been applied to the gap-filling radar sensor, while for Figure 3 In the vehicle 120 shown in (c) and (d), the gap-filling radar sensor installed on the side of the vehicle 120 has a yaw angle of -20° and +20°, respectively.

[0095] exist Figure 4 4 shows in detail coverage areas 410 to 478 of the individual radar sensors installed on vehicle 120 . Figure 4 (a) corresponds to Figure 3 The situations shown in (a) and (b) of FIG. 1 , where the entire coverage area of ​​the radar sensor is not optimized at the rear axle of the vehicle 120 as required for low-speed and parking applications. Instead, Figure 4 (b) corresponds to Figure 3 The situations shown in (c) and (d) of FIG. 120 , wherein the overall coverage area of ​​the radar sensor is optimized at the rear axle by applying corresponding yaw angles of −20° and +20°, respectively, to the gap-filling radar sensors mounted on the sides of the vehicle 120 . Figure 4In (a) and (b) of FIG. 4 , coverage areas 410 to 478 at a height of 30 cm relative to the ground plane are shown. The plane at a height of 30 cm above the ground plane 210 is also Figure 2 For this plane, a blind spot distance of approximately 14 cm is calculated, which is also shown in Figure 4 is shown in (a).

[0096] In detail, Figure 4 (a) and (b) of FIG. 1 show a top view of the vehicle and the corresponding coverage areas provided by different radar sensors installed on the vehicle 120, respectively. Figure 4 In (a) and (b), four main radar sensors are installed at the corners of the vehicle 120 at a mounting height of 50 cm. The respective coverage areas of the main radar sensors are Figure 4 410, 420, 430 and 440 in (a) and (b). The right front radar sensor is installed at a yaw angle of 60°, while the left front radar sensor is installed at a yaw angle of -60°. For the front right radar sensor and the front left radar sensor, the corresponding coverage areas are represented by 410 and 420, respectively. In contrast, the right rear radar sensor and the left rear radar sensor are installed at yaw angles of 45° and -45°, respectively. The coverage areas of the right rear radar sensor and the left rear radar sensor are represented by 430 and 440, respectively.

[0097] In addition, Figure 4 The coverage areas of four gap filler radar sensors are also shown in (a) and (b) of FIG. 1. The gap filler radar sensors are mounted on the sides of the vehicle 120 between the main radar sensors located at the corners of the vehicle 120. The four gap filler radar sensors are mounted at a pitch angle of -45°.

[0098] The front gap-filling radar sensor is mounted at a mounting height of 70 cm, and the corresponding coverage area is denoted by 450. The rear gap-filling radar sensor is mounted at a mounting height of 90 cm, and its coverage area is denoted by 460.

[0099] The gap-filling radar sensor mounted on the side of the vehicle is installed at a mounting height of 100 cm. Figure 4 In the case shown in (a), the gap-filling radar sensor mounted on the side of the vehicle has a yaw angle of 0°. Figure 4 In (a), the corresponding coverage areas of the side gap-filling radar sensors are represented by 472 and 474, respectively.

[0100] On the contrary, for Figure 4 In the case shown in (b), the side gap filler radar sensors are installed at yaw angles of -20° and +20° respectively. The corresponding coverage areas of these side gap filler sensors are Figure 4In (b), they are represented by 476 and 478 respectively.

[0101] As in Figure 4 As can be appreciated in (a) of FIG. 1 , a corresponding blind spot volume 320 represented as a shaded area at a height of 30 cm relative to the ground plane and not covered by any radar sensor mounted on the vehicle 120 appears close to the rear axle of the vehicle 120. In contrast, by Figure 4 With the radar sensor installation shown in (b), the blind spot volume 320 is greatly reduced, while the blind spot volume 330 at the front axle of the vehicle is increased. Figure 4 Optimization of the coverage area near the rear axle of the vehicle 120 shown in b) allows for correct execution of low-speed and parking applications.

[0102] Figure 5 A flow chart 500 illustrating a method of positioning perception sensors on a vehicle is shown. At 502, a three-dimensional coverage map of the perception sensors can be determined. At 504, a vehicle-specific geometry can be received. At 506, a coverage area of ​​the perception sensors near the vehicle can be estimated by combining the vehicle-specific geometry and the three-dimensional coverage map of the perception sensors and by changing the spatial position of the perception sensors until the coverage area near the vehicle is optimized.

[0103] According to various embodiments, changing the spatial position of the perception sensor may include changing the positioning height of the perception sensor relative to the ground plane and changing the yaw, pitch, and roll angles of the perception sensor relative to the vehicle.

[0104] According to various implementations, the perception sensor may include a radar sensor, and determining the three-dimensional coverage map may include applying a link budget equation and a predetermined radar cross section.

[0105] According to various embodiments, an optimized coverage area may be determined by using a predetermined fundamental frequency of the radar sensor in order to determine an optimized spatial position of the radar sensor. A corresponding three-dimensional coverage map may be determined for each of a predetermined number of frequencies of the radar sensor that are different from the fundamental frequency, and a corresponding coverage area of ​​the radar sensor may be estimated in the vicinity of the vehicle for each of the predetermined number of frequencies of the optimized spatial position of the radar sensor by applying the corresponding three-dimensional coverage map.

[0106] According to various embodiments, the vehicle-specific geometry may include a CAD model of the vehicle.

[0107] According to various embodiments, shadow areas in the predetermined area of ​​interest that are outside the coverage area may be determined by taking into account the geometry provided by the CAD model of the vehicle.

[0108] According to various embodiments, at least one blind spot volume located outside the coverage area may be determined relative to a CAD model of the vehicle.

[0109] Each of steps 502, 504, 506, and the additional steps described above, may be performed by computer hardware components.

[0110] Figure 6 A perception sensor positioning system 600 is shown according to various embodiments. The perception sensor positioning system 600 may include a coverage map determination circuit 602, a vehicle geometry reception circuit 604, and a coverage area estimation circuit 606, which may include a combination circuit 608 and an alteration circuit 610.

[0111] Coverage map determination circuit 602 may be configured to determine a three-dimensional coverage map of the perception sensor. Vehicle geometry receiving circuit 604 may be configured to receive vehicle-specific geometry characteristics. Coverage area estimation circuit 606 may be configured to estimate the coverage area of ​​the perception sensor near the vehicle by using combining circuit 608 and changing circuit 610. Combining circuit 608 may be configured to combine the vehicle-specific geometry characteristics of the perception sensor and the three-dimensional coverage map. Changing circuit 610 may be configured to change the spatial position of the perception sensor until the coverage area near the vehicle is optimized.

[0112] Coverage map determination circuit 602, vehicle geometry reception circuit 604, coverage area estimation circuit 606, combination circuit 608 and variation circuit 610 may be coupled to one another, for example, via an electrical connection 611 such as a cable or a computer bus, or via any other suitable electrical connection to exchange electrical signals.

[0113] A “circuit” may be understood as any type of logic-implemented entity, which may be a dedicated circuit or a processor executing a program stored in a memory, firmware, or any combination thereof.

[0114] Figure 7 A computer system 700 having multiple computer hardware components configured to perform the steps of a computer-implemented method for integrating a radar sensor in a vehicle according to various embodiments is shown. The computer system 700 may include a processor 702, a memory 704, and a non-transitory data storage device 706.

[0115] The processor 702 may execute instructions provided in the memory 704. The non-transitory data storage 706 may store a computer program including instructions that may be transferred to the memory 704 and then executed by the processor 702.

[0116] The processor 702 , memory 704 , and non-transitory data storage device 706 may be coupled to one another, for example, via an electrical connection 708 , such as a cable or a computer bus, or via any other suitable electrical connection to exchange electrical signals.

[0117] As such, processor 702, memory 704, and non-transitory data storage 706 may represent coverage map determination circuit 602, vehicle geometry reception circuit 604, coverage area estimation circuit 606, combination circuit 608, and variation circuit 610, as described above.

[0118] The terms “coupled” or “connected” are intended to include a direct “coupled” (eg, via a physical link) or direct “connected” as well as an indirect “coupled” or indirect “connected” (eg, via a logical link), respectively.

[0119] It should be understood that what has been described with respect to one of the above methods may similarly hold true for the perception sensor positioning system 600 and / or for the computer system 700 .

Claims

1. A computer-implemented method for positioning a perception sensor (110) on a vehicle (120), the method comprising the steps of: Determining a three-dimensional coverage map (100) of the perception sensor (110); receiving a vehicle-specific geometric characteristic; as well as The coverage area (476, 478) near the vehicle (120) is estimated by combining the vehicle-specific geometric features and the three-dimensional coverage map (100) of the perception sensor (110) and by changing the spatial position of the perception sensor (110) until the coverage area (476, 478) of the perception sensor (110) near the vehicle (120) is optimized.

2. The computer-implemented method of claim 1 , wherein: The step of changing the spatial position of the perception sensor (110) includes: changing the positioning height of the perception sensor (110) relative to the ground plane (210), and changing the yaw angle, pitch angle and roll angle of the perception sensor (110) relative to the vehicle (120).

3. The computer-implemented method of claim 1 or 2, wherein: The perception sensor (110) includes a radar sensor, and determining the three-dimensional coverage map (100) includes applying a link budget equation and a predetermined radar cross section.

4. The computer-implemented method of claim 3, wherein: The optimized coverage area (476, 478) is determined by using a predetermined fundamental frequency of the radar sensor (110) to determine an optimized spatial position of the radar sensor (110), A corresponding three-dimensional coverage map is determined for each frequency of a predetermined number of frequencies of the radar sensor (110) that are different from the fundamental frequency, and A corresponding coverage area of ​​the radar sensor (110) is estimated in the vicinity of the vehicle (120) by applying the corresponding three-dimensional coverage map for each of the predetermined number of frequencies for the optimized spatial position of the radar sensor (110).

5. The computer-implemented method of claim 1 , wherein: The vehicle-specific geometry includes a CAD model of the vehicle (120).

6. The computer-implemented method of claim 5, wherein: By considering the geometry provided by the CAD model (200) of the vehicle (120), a shadow region (230) in a predetermined region of interest outside of the coverage area (476, 478) is determined.

7. A computer-implemented method according to claim 5 or 6, wherein: At least one blind spot volume (320) located outside of the coverage area (476, 478) is determined relative to the CAD model (200) of the vehicle (120).

8. A computer system configured to perform the computer-implemented method according to any one of claims 1 to 7.

9. A vehicle (120), comprising a perception system having at least one perception sensor (110), wherein: The spatial position of the perception sensor (110) relative to the vehicle (120) has been determined by executing a computer-implemented method according to any one of claims 1 to 7.

10. The vehicle (120) of claim 9, wherein: The spatial position of the at least one perception sensor (110) is determined such that at least one blind spot volume (320) within a desired coverage area of ​​the perception sensor (110) is optimized at the vehicle (120).

11. The vehicle (120) according to claim 9 or 10, wherein: The perception system includes at least two perception sensors (110), the at least two perception sensors having overlapping three-dimensional coverage areas.

12. The vehicle (120) according to claim 9 or 10, wherein: The perception system includes a radar system including at least one radar sensor (110).

13. The vehicle (120) of claim 12, wherein: The radar system includes at least one front radar sensor, at least one radar sensor located on a side of the vehicle, and at least one rear radar sensor.

14. The vehicle (120) according to claim 9 or 10, wherein: The perception system includes a lidar system, which includes at least one lidar sensor.

15. A non-transitory computer-readable medium comprising instructions for executing the computer-implemented method according to any one of claims 1 to 7.

Images