Method for operating a positioning system of a motor vehicle and positioning system

Abstract

The invention relates to a method for operating a positioning system (2) of a motor vehicle (1), in which method a low-dimensional first pose (P1) of the motor vehicle (1) is determined in a two-dimensional environment (6) of the motor vehicle (1) by means of a first positioning device (3) of the positioning system (2), and in which method a higher-dimensional second pose (P2) of the motor vehicle (1) is determined in a three-dimensional environment (6) of the motor vehicle (1) by means of a second positioning device (4) of the positioning system (2), which is separate from the first positioning device (3). According to the invention, for determining the higher-dimensional second pose (P2), the low-dimensional first pose (P1) of the first positioning device (3) is transmitted to the second positioning device (4) and the second pose (P2) is determined therefrom. The invention also relates to a positioning system (2).

Description

TECHNICAL FIELD

[0001] The invention relates to a method for operating a positioning system of a motor vehicle, in which method a first, low-dimensional attitude of the motor vehicle is determined in a two-dimensional environment of the motor vehicle by means of a first positioning device of the positioning system, and in which method a second, higher-dimensional attitude of the motor vehicle is determined in a three-dimensional environment of the motor vehicle by means of a second positioning device of the positioning system separate from the first positioning device. Furthermore, the invention relates to a computer program product and to a positioning system. BACKGROUND

[0002] Driving functions, for example to assist a driver of a motor vehicle, are known in motor vehicle construction. Furthermore, display, assistance, driving and highly automated driving functions are also known, which in turn require information about the position and orientation (attitude) of the motor vehicle. This can be provided by a vehicle-own positioning system. Depending on the driving function, there can be hard or soft requirements for the accuracy and range of the attitude data provided by the positioning system. Highly automated driving functions, for example, require all three dimensions of the position (x-dimension, y-dimension, z-dimension) and all three angles of the orientation (pitch angle, roll angle, yaw angle). It is also important here what circumstances should be used for the function. Functions on motorways and local roads, for example, can dispense with z-information and instead assume that the motor vehicle is at the respective position on the "earth's surface". Functions for assistance in a multi-storey car park, on the other hand, require information in the z-dimension, i.e. the height, in order to be able to describe a multi-storey car park, for example.

[0003] The patent document US 10,782,411 B2 describes a system and method for determining a position of a vehicle. A first positioning device is capable of producing a first attitude estimate for the vehicle based at least in part on a comparison of first remote sensor data to first reference data. A second positioning device is capable of producing a second attitude estimate for the vehicle based at least in part on a comparison of second remote sensor data to second reference data. A position state estimation device can produce a vehicle position for the vehicle based at least in part on a first previous position of the vehicle, the first position estimate, and the second position estimate.

[0004] Patent document US 9,342,888 B2 describes a system and method for mapping, localization and position correction, comprising determining a current position of a vehicle along a travel route and determining a set of currently observed landmarks along the travel route relative to the current position, wherein the set of currently observed landmarks is extracted from one or more stereoscopic images obtained by an imaging device, further comprising retrieving a measured landmark database in order to validate a part or subset of the set of measured landmarks relative to the current position of the vehicle. The method comprises determining one or more two-dimensional transformation estimates between the set of currently observed landmarks and the subset of measured-related landmarks and determining a best transformation estimate from the one or more two-dimensional transformation estimates, which best transformation estimate minimizes a distance between the set of currently observed landmarks and the subset of measured-related landmarks. The method includes correcting a pose of the vehicle based on the best transformation estimate.

[0005] Patent document US 2020 / 0 217 972 A1 describes a method for vehicle localization, the method comprising determining a first 6-DOF pose of a vehicle, wherein the first 6-DOF pose can comprise a first height and one or more first rotation parameters, the one or more first rotation parameters showing a first orientation of the vehicle relative to a reference frame. A lane plane related to a road surface on which the vehicle travels can be determined from the first 6-DOF pose and positions of lane boundary markers on the road surface. A respective position of a lane boundary marker can be determined from a map for each lane boundary marker, the map can be based on the reference frame. A corrected height of the vehicle can be determined from the lane plane. A corrected 6-DOF position of the vehicle can be determined from the corrected height of the vehicle, the first 6-DOF position and an axis perpendicular to the road surface plane.

[0006] Patent document US 2020 / 0 142 074 A1 discloses a method for position calculation for a portable three-dimensional scanning device, the three-dimensional scanning device comprising a first sensor and a second sensor, wherein the method comprises using data from the first sensor and data from the second sensor to obtain data defining six degrees of freedom of the scanning device in order to optimize a first position calculation, wherein receiving data is performed, the data comprising both data from the first sensor and data from the second sensor, selecting a subset of the six degrees of freedom of the scanning device, optimizing a second pose using data from the first sensor and the received data for the selected subset of the six degrees of freedom, wherein the unselected degrees of freedom of the first pose are retained, and storing the received data related to the second camera pose in a point cloud database.

[0007] The patent document DE 10 2020 213 133 A1 describes an apparatus for performing a localization by means of sensor data, having: - a first determination apparatus for determining a first pose from first sensor data; - a second determination apparatus for determining a second pose from second sensor data; and - a verification apparatus for verifying the first pose determined by the first determination apparatus, having: - a first comparison apparatus for comparing the first pose and the second pose, wherein a confidence signal relating to a confidence of the performed self-localization can be output in accordance with the comparison performed by means of the verification apparatus. SUMMARY

[0008] The technical problem addressed by the present application is to provide a method, a computer program product and a localization system by means of which the current position or the current pose of a motor vehicle in an environment of the motor vehicle can be determined more improved.

[0009] The technical problem is solved according to the present application by a method for operating a localization system of a motor vehicle, a computer program product and a localization system.

[0010] One aspect of the present application relates to a method for operating a localization system of a motor vehicle, in which a low-dimensional first pose of the motor vehicle in a two-dimensional environment of the motor vehicle is determined by means of a first localization apparatus of the localization system, and in which a higher-dimensional second pose of the motor vehicle in a three-dimensional environment of the motor vehicle is determined by means of a second localization apparatus of the localization system separate from the first localization apparatus, wherein the higher-dimensional second pose is higher-dimensional at least relative to the first pose.

[0011] It is provided that, for determining the higher-dimensional second pose, the low-dimensional first pose of the first localization apparatus is transmitted to the second localization apparatus and the second pose is determined therefrom.

[0012] This makes it possible to determine the position or the pose of the motor vehicle in the environment more improved. The localization system here makes use of, inter alia, the strengths of the localization of the low-dimensional first pose and the determination of the higher-dimensional second pose and combines them again.

[0013] What is used here in particular is that, for cases in which only a small amount of data is required, i.e. in particular only a low-dimensional pose is required, the mathematical complexity required for estimating the vehicle pose is far lower than for cases in which all six degrees of freedom, i.e. a higher-dimensional pose, are required. Depending on the driving function in which the motor vehicle is located, it is possible for the localization system to use either the first localization apparatus or the second localization apparatus, which can be decided in accordance with the requirements of the driving function.

[0014] The combination of the positioning information from the first positioning device and the second positioning device is advantageous, inter alia, for reasons of robustness, availability and the field of application. The first positioning device and the second positioning device can use, for example, GPS (Global Positioning System), odometry data or prominent landmarks identified by, for example, a camera. The mathematical complexity of the combination scheme depends, as described, to a large extent on the required attitude data. Especially in parking spaces in multi-storey car parks, the combination scheme must also take into account the z-dimension. In addition, highly automated driving functions, for example Automated Valet Parking, can require consideration of all three position angles.

[0015] The system thereby uses the coupling of a low-dimensional first positioning device with a high-dimensional, fine second positioning device, with the advantage that the strengths of both positioning devices are exploited and the respective weaknesses are compensated.

[0016] The second positioning device therefore, inter alia, uses the first attitude as an initial attitude assumption and thereby initialises its own algorithm. The second positioning device can now reduce the expenditure for estimating the missing dimension and create and output, for example, a second attitude, so that the subsequent assistance system, for example for automated driving, obtains attitude data of high dimensionality.

[0017] The low-dimensional first attitude is, for example, inter alia, a two-dimensional attitude or the first attitude can be described, for example, by at least two parameters (X-dimension, Y-dimension) or three parameters or at least three degrees of freedom. The second attitude of higher dimensionality is, inter alia, of higher dimensionality than the first attitude, the second attitude can be, for example, three-dimensional and / or can be described, for example, by six degrees of freedom. This is purely exemplary. However, it is important that the second attitude has a higher dimensionality than the first attitude. The number of dimension differences is not important here. It is also possible that the first attitude can be described, for example, by four degrees of freedom, wherein the second attitude then has at least more than four degrees of freedom. It is noted here that fractal dimensionality can also be taken into account when determining the attitude.

[0018] According to an advantageous design, the second attitude is transmitted to the first positioning device after determination of the second attitude and is taken into account when determining the future first attitude. The first positioning device can thereby use the higher quality and more precise attitude information of the second attitude of the second positioning device in order to itself use the more precise information for the future calculation and thus already be able to determine the first attitude more precisely within the first positioning device.

[0019] It is furthermore advantageous to determine a first pose having an x-dimension of the motor vehicle and having a y-dimension of the motor vehicle and having a yaw angle of the motor vehicle. Thus, a two-dimensional position in the environment is determined, in particular. Here, a low-dimensional pose is meant, which can be used to provide corresponding driving functions, for example, on a motorway or on a local road. Here, higher-quality pose data is not required, because additional information can be used or because these driving functions do not require higher-quality functions.

[0020] In another advantageous design, a second pose is determined having an x-dimension of the motor vehicle and having a y-dimension of the motor vehicle and having a z-dimension of the motor vehicle and having a yaw angle of the motor vehicle and having a roll angle of the motor vehicle and having a pitch angle of the motor vehicle. Here, the position of the motor vehicle is determined using all six degrees of freedom, in particular. This is necessary, for example, in a corresponding three-dimensional garage drive, in particular in an automated three-dimensional garage drive, in order to be able to determine the exact position of the motor vehicle within the three-dimensional garage. Here, it is necessary to recognize, for example, that it is also decisive which level the motor vehicle is on, so that the z-dimension is used, which also corresponds to the height of the motor vehicle above the earth's surface. The x-dimension is meant, in particular, the longitudinal dimension and the y-dimension is meant the width dimension.

[0021] In another advantageous design, at least the first pose is determined on the basis of satellite navigation and / or on the basis of detection of landmarks by means of a camera of a positioning system and / or on the basis of detection of the environment by means of a radar sensor device and / or on the basis of detection of the environment by means of a lidar sensor device and / or on the basis of odometer data of the motor vehicle. Thus, the low-dimensional first positioning device provides, for example, dimensions, in particular x-dimension, y-dimension and yaw angle, on the basis of pose data, and for this purpose includes a plurality of positioning sources, for example, global positioning system (GPS), odometer data, camera landmarks and radar sensor data and lidar sensor data. Here, the first pose can be determined in a simple manner.

[0022] According to another advantageous design, the first pose is determined at least by means of a graph-based optimization method. The combination scheme for determining the first pose is thus based, in particular, on a graph-based optimization method. Here, a factor graph having the necessary dimensions is created and an iterative optimization method is applied, which usually requires matrix operations. Here, the fewer the dimensions required, the better the method can be implemented, for example, on the motor vehicle hardware and / or on the motor vehicle software.

[0023] It is furthermore advantageous if the first attitude is transmitted from the first positioning device to the second positioning device zyklisch, that is to say cyclically, and / or at preset time intervals. Thereby, inter alia, the feasibility of how the first positioning device and the second positioning device can be coupled to one another is shown, for example, in that the first attitude is transmitted to the second positioning device zyklisch, that is to say according to a determined time pattern, that is to say at preset time intervals, and enriched with its own data. Thereby, a calculation step within the second positioning device is saved by using the first attitude.

[0024] It is furthermore proven to be advantageous if the first attitude is adapted according to the second attitude and the first attitude and the second attitude are output by means of the second positioning device. This can be output, for example, to a functional device of the motor vehicle, for example to an assistance system of the motor vehicle. Thereby, inter alia, it is possible for the first attitude as well as the second attitude to be output by means of the second positioning device, wherein the first attitude has here component data of higher quality and can thus be considered as an additional source for positioning information, for example for an assistance system, and can thus also be adapted accordingly.

[0025] The proposed method is a computer-implemented method. Therefore, another aspect of the present application relates to a computer program product having program code instructions which, when executed by an electronic computing device, cause the electronic computing device to perform the method according to the preceding aspect. Therefore, another aspect of the present application also relates to a computer-readable storage medium having the computer program product.

[0026] Furthermore, the present application also relates to a positioning system for a motor vehicle, having at least one first positioning device and having a second positioning device separate from the first positioning device, wherein the positioning system is designed to perform the method according to the preceding aspect. The method is performed, inter alia, by means of the positioning system.

[0027] The positioning system has here, inter alia, at least one electronic computing device, which can have, for example, a processor, a circuit, inter alia, an integrated circuit, and other components, in order to be able to perform the respective method steps.

[0028] Furthermore, the present application also relates to a motor vehicle having a positioning system according to the preceding aspect. The motor vehicle is here, inter alia, designed to be at least partially automatic or fully automatic.

[0029] The present application also comprises an extended design of the positioning system according to the present application, which has the features already described in connection with the extended design of the method according to the present application. Therefore, the respective extended design of the positioning system according to the present application is not described here again in detail.

[0030] The application also comprises combinations of features of the described embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0031] Embodiments of the application are described below. In this connection, reference is made to the accompanying drawings, in which:

[0032] Figure 1 a schematic top view of an embodiment of a motor vehicle with an embodiment of a positioning system is shown; and

[0033] Figure 2 a schematic block diagram according to an embodiment of a positioning system is shown.

[0034] The embodiments described below are preferred embodiments of the application. In this embodiment, the components described are each individual, mutually independent features of the application, which also each form an extension of the application independently of one another and can therefore also be considered constituent parts of the application, either individually or in a combination different from the one shown. Furthermore, the embodiments described can also be supplemented by other features of the application already described.

[0035] In the drawings, functionally identical elements are each provided with the same reference symbols. DETAILED DESCRIPTION

[0036] Figure 1 a schematic top view of an embodiment of a motor vehicle 1 with an embodiment of a positioning system 2 is shown. The positioning system 2 has at least one first positioning device 3 and a second positioning device 4. Furthermore, the positioning system 2 can also have an electronic computing device 5. In a method for operating the positioning system 2, a low-dimensional first pose P1 of the motor vehicle 1 in a two-dimensional environment 6 of the motor vehicle 1 is determined by means of the first positioning device 3, and a higher-dimensional second pose P2 of the motor vehicle 1 in a three-dimensional environment 6 of the motor vehicle 1 is determined by means of the second positioning device 4, which is separate from the first positioning device 3.

[0037] It is provided here that, for determining the higher-dimensional second pose P2, the low-dimensional first pose P1 of the first positioning device 3 is transmitted to the second positioning device 4 and the second pose P2 is determined from the first pose P1.

[0038] Figure 1 It is also shown that at least the first pose P1 has an x-dimension 7 of the motor vehicle 1 and has a y-dimension 8 of the motor vehicle 1 and has a yaw angle of the motor vehicle 1. Furthermore, the second pose P2 has an x-dimension 7 of the motor vehicle 1 and has a y-dimension 8 of the motor vehicle 1 and has a z-dimension 9 of the motor vehicle 1 and has a yaw angle of the motor vehicle 1 and has a roll angle of the motor vehicle 1 and has a pitch angle of the motor vehicle 1.

[0039] It is furthermore shown that at least the first attitude Pi can be determined on the basis of satellite navigation and / or on the basis of the detection of landmarks by means of the detection device 10 of the motor vehicle 1 or of the positioning system 2, wherein, as detection device 10, here for example a video camera and / or a radar sensor device and / or a lidar sensor device and / or an ultrasonic sensor device can be used. The first attitude Pi and / or the second attitude P2 can furthermore be determined on the basis of the odometer data of the motor vehicle 1.

[0040] Figure 2 A schematic block diagram is shown in accordance with an embodiment of the positioning system 2. It is shown here, inter alia, that the first attitude Pi is transmitted by the first positioning device 3 to the second positioning device 4. The second positioning device 4 in turn processes the first attitude Pi on the basis of the first attitude Pi and generates the second attitude P2. The second attitude P2 can then for example also be transmitted to the assistance system 11 of the motor vehicle 1 for further driving functions.

[0041] Figure 2 It is furthermore shown that the second attitude P2 can be transmitted to the first positioning device 3 after determination of the second attitude P2 and that the second attitude P2 can be taken into account when determining the future first attitude Pi. It is furthermore shown that the first attitude Pi can be adjusted to Pi' on the basis of the second attitude P2 here and that the adjusted first attitude Pi' and the second attitude P2 can be output by means of the second positioning device 4 and for example transmitted to the assistance system 11 as shown here. It can furthermore be provided that the first attitude Pi is transmitted from the first positioning device 3 to the second positioning device 4 cyclically and / or at preset time intervals. It can furthermore be provided that the first attitude Pi is determined by means of a chart-based optimization method.

[0042] Thus, Figure 2 It is shown in particular that a first positioning device 3 of low dimensionality is coupled with a second positioning device 4 of high dimensionality, which has the advantage that the strengths of both positioning devices 3, 4 are fully utilized and the respective weaknesses are compensated.

[0043] The first positioning device 3 provides, for example, dimensions for the x dimension 7, the y dimension 8 and the yaw angle on the basis of the attitude data and for this purpose includes a plurality of positioning sources, for example a global satellite system, odometry data, landmarks detected by a camera and radar and lidar. The strength of the first positioning device 3 is that, due to the lower dimensionality, there is lower mathematical complexity for the calculation, combination and finally for the attitude data of the intermediate results. Current combination concepts are, for example, usually based on graph-based optimization methods. Here, a factor graph with the necessary dimensions is created and an iterative optimization method is applied, which usually requires matrix operations. Here, the fewer dimensions that are required, the better these methods can be implemented on the motor vehicle hardware and / or motor vehicle software. However, the first attitude PI can, for example, also not be used for higher-quality driving functions which require highly precise attitude data.

[0044] The second positioning device 4 provides, for example, attitude data with all six dimensions. According to the application, however, the second positioning device 4 no longer performs a complete positioning, but rather builds on the first attitude data of the first attitude PI of the first positioning device 3. The possibility of coupling the two positioning devices 3, 4 lies in the fact that the second positioning device 4 obtains the first attitude PI cyclically or according to a determined time pattern and enriches it with its own data. By using the first attitude PI, the calculation step is advantageously saved. The enrichment can, for example, manifest itself in the fact that the second positioning device 4 uses the first attitude PI as an initial attitude assumption and initializes its own algorithm therefrom in order to produce the second attitude P2. The missing dimensions can now be estimated, created and output with lower outlay.

[0045] It can furthermore be provided that the first positioning device 3 uses the higher-quality and more precise attitude data of the second attitude P2 in order to optimize the calculation itself. For this purpose, the first positioning device 3 extracts those data from the second attitude P2 which correspond to the format of the first attitude PI. As an alternative or in addition, the second positioning device 4 can output, in addition to the second attitude P2, the first attitude PI, here in particular the adjusted first attitude PI', with higher-quality components data. The first positioning device 3 can thereby use the second attitude P2 as a further source of positioning information and combine it accordingly.

Claims

1. A method for a positioning system (2) for operating a motor vehicle (1), wherein a low-dimensional first pose (P1) of the motor vehicle (1) is determined in a two-dimensional environment (6) of the positioning system (2) by means of a first positioning device (3) of the positioning system (2), and a higher-dimensional second pose (P2) of the motor vehicle (1) is determined in a three-dimensional environment (6) of the motor vehicle (1) by means of a second positioning device (4) of the positioning system (2) separate from the first positioning device (3), wherein, The higher-dimensional second pose (P2) is at least higher in dimension than the first pose (P1). Its features are, In order to determine the higher-dimensional second pose (P2), the lower-dimensional first pose (P1) of the first positioning device (3) is transmitted to the second positioning device (4) and the second pose (P2) is determined based on the first pose (P1).

2. The method according to claim 1, Its features are, After determining the second posture (P2), the second posture (P2) is transmitted to the first positioning device (3) and the second posture (P2) is taken into account when determining the future first posture (P1).

3. The method according to claim 1 or 2, Its features are, A first posture (P1) is determined having the x dimension (7) of the motor vehicle (1) and the y dimension (8) of the motor vehicle (1) and the deflection angle of the motor vehicle (1).

4. The method according to claim 1 or 2, Its features are, A second posture (P2) is determined, having the x dimension (7) of the motor vehicle (1), the y dimension (8) of the motor vehicle (1), the z dimension (9) of the motor vehicle (1), the yaw angle of the motor vehicle (1), the roll angle of the motor vehicle (1), and the tilt angle of the motor vehicle (1).

5. The method according to claim 1 or 2, Its features are, At least the first posture (P1) is determined based on satellite navigation and / or based on the detection of landmarks by a camera using the positioning system (2) and / or based on the detection of the environment (6) using a radar sensor device and / or based on the detection of the environment (6) using a lidar sensor device and / or based on the odometer data of the vehicle (1).

6. The method according to claim 1 or 2, Its features are, At least the first pose (P1) is determined using a graph-based optimization method.

7. The method according to claim 1 or 2, Its features are, The first posture (P1) is transmitted from the first positioning device (3) to the second positioning device (4) cyclically and / or at preset time intervals.

8. The method according to claim 1 or 2, Its features are, The first posture (P1) is adjusted according to the second posture (P2), and the adjusted first posture (P1′) and second posture (P2) are output by means of the second positioning device (4).

9. A computer program product having program code instructions, wherein when the program code instructions are executed by an electronic computing device (5), the program code instructions cause the electronic computing device (5) to perform the method according to any one of claims 1 to 8.

10. A positioning system (2) for a motor vehicle (1), the positioning system having at least one first positioning device (3) and a second positioning device (4) separate from the first positioning device (3), wherein, The positioning system (2) is designed to perform the method according to any one of claims 1 to 8.

Images