Method for creating generally applicable feature maps

Abstract

A method for creating a digital map by a control device is disclosed, wherein during a measurement drive, measurement data of the surroundings are received, a SLAM method is performed based on the received measurement data to determine a trajectory of the measurement drive, the received measurement data are converted into a coordinate system of the trajectory, the converted measurement data are used to create an intensity map, features are extracted from the intensity map and stored in a feature map. In addition, a method for performing localization, a control device, a computer program, and a machine-readable storage medium are disclosed.

Description

Technical Field

[0001] The invention relates to a method for creating a digital map and a method for performing a positioning operation. The invention also relates to a control device, a computer program and a machine-readable storage medium. Background Art

[0002] For the automated operation of vehicles and robots, positioning is an indispensable functional component. Through positioning, the exact position of the vehicle or robot in a map or in the surrounding environment can be determined. Based on the determined position, control commands can be generated so that, for example, a trajectory is traveled or a task is executed.

[0003] In particular, when using without access to GNSS data, so-called SLAM methods are used for simultaneous positioning and mapping. For this purpose, measurement data are collected from, for example, a lidar sensor and analyzed to generate a map. In a subsequent step, the position within the map can be determined.

[0004] The application of SLAM methods in dynamic or semi-static environments is problematic. Such environments may exist, for example, in warehouses, construction sites, intralogistics or container ports. Due to the regular movement of objects, the created maps lose their validity for a short time. Regular updating of such maps requires a high measurement and analysis overhead. In particular, all users must be provided with regularly updated maps, which makes it necessary to provide an infrastructure for providing a high amount of data. Summary of the invention

[0005] The object on which the present invention is based can be seen as providing a method for creating a generally applicable digital map with reduced data consumption.

[0006] This object is achieved by means of the respective subject matter of the independent claims. Advantageous embodiments of the invention are the subject matter of the corresponding dependent claims.

[0007] According to one aspect of the invention, a method for creating a digital map by a control device is provided. In one step, measurement data of the surroundings are received during a measurement drive. The measurement drive can be any drive. Preferably, the measurement data can be collected by at least one sensor even when the vehicle is stopped or parked. The corresponding measurement data can then be received and processed by the control device.

[0008] Based on the received measurement data, a SLAM method is performed to determine the trajectory of the measurement run. Here, self-positioning is performed based on a series of measurement data, wherein the corresponding positions during the measurement run form the trajectory.

[0009] In a further step, the received measurement data are converted into a coordinate system of the trajectory. The received measurement data may, for example, have relative positions and / or distances relative to the sensor. These relative coordinates may then, for example, be converted into an absolute coordinate system of the trajectory. Such a coordinate system may, for example, be a Cartesian coordinate system.

[0010] The converted measurement data are used to create an intensity map. For example, the intensity of the reflected radiation of one or more lidar sensors or radar sensors can be determined and stored in the form of a map with the received radiation intensities.

[0011] Features are then extracted from the intensity map and stored in a feature map. These features can preferably be detected in the intensity map. This process can be performed, for example, by an algorithm for pattern recognition. Pattern recognition can also be performed by a previously correspondingly trained neural network. Pattern recognition can be performed, for example, manually or automatically by an operator. In addition, the pattern recognition performed automatically can be authorized or confirmed by the operator.

[0012] The pattern map can preferably be universally applicable. In particular, the pattern map can be used sensor-independently or across sensors, so that features can be extracted from differently ascertained measurement data and can be used for localization according to the feature map.

[0013] According to another aspect of the invention, a method for performing positioning, in particular by means of a control device, is provided. In one step, measurement data and a feature map of the surroundings are received. The measurement data can be obtained by one or more sensors. Such sensors can be, for example, camera sensors, lidar sensors, radar sensors, ultrasonic sensors, etc. The sensor can in particular be different from the sensor used to create the feature map.

[0014] In a further step, features are identified and extracted from the received measurement data. At least one extracted feature is then compared with features stored in a feature map to determine the position. If at least one feature is successfully compared, the position of the sensor or of the vehicle that performs the measurement with the aid of the sensor can be determined.

[0015] According to another aspect of the invention, a control device is provided which is configured to perform the method. Here, the control device may be a vehicle-internal control device which is integrated into a vehicle control device for implementing automated driving functions or may be connected to the vehicle control device. Alternatively or additionally, the control device may be configured as a vehicle-external control device, such as a server unit or cloud technology.

[0016] In addition, according to one aspect of the present invention, there is provided a computer program comprising instructions, which, when executed by a computer or a control device, cause the computer or the control device to perform the method according to the present invention. According to another aspect of the present invention, there is provided a machine-readable storage medium on which a computer program according to the present invention is stored.

[0017] In this case, the control device can be mounted on a vehicle. In particular, at least one measurement run can be carried out in a vehicle having the control device. The vehicle can be operated in an assisted, partially automated, highly automated and / or fully automated or driverless manner according to the BASt (Bundesanstalt für Straβenwese, German Federal Institute for Traffic Research) standard. According to an alternative or additional configuration, the vehicle can be a robot, a drone, a water vehicle or the like. As a result, the method can be used on the street, for example on motorways, highways, urban areas, and off the street or in the wild (offroad). In particular, the method can be used in buildings or halls, underground spaces, parking lots and garages, tunnels, etc.

[0018] The at least one sensor for determining the measurement data may be a component of the surrounding sensor system or at least one sensor of the vehicle. The at least one sensor may in particular be a lidar sensor, a radar sensor, an ultrasonic sensor, a camera sensor, an odometer, an acceleration sensor, a position sensor, etc. These sensors may in particular be used individually or in combination with one another. Furthermore, sensors such as acceleration sensors, wheel sensors, lidar sensors, ultrasonic distance sensors, cameras, etc. may also be used to perform the odometry method.

[0019] In particular, static features of the surrounding environment can be obtained and extracted by the method according to the invention. Such features can be, for example, lane markings, the geometry of buildings, curbs, streets, the arrangement and position of traffic lights, guide posts, lane boundaries, buildings, containers, etc. Such features can be detected by different sensors and used for positioning. Features extracted from the measurement data of a lidar sensor can also be detected, for example, by a camera sensor and compared with each other for the purpose of positioning. It is thus possible to create a universally applicable feature map that can be used by different vehicles and machines. For example, it is conceivable that such feature maps can be used for precise positioning by manned spacecraft, transport units, manipulators, etc.

[0020] Preferably, in a first step, a marking map can be created and subsequently used for localization tasks. The marking map can be used in particular for localization tasks and control tasks of an automatically operated vehicle or robot.

[0021] Since the features of the feature map can exist as geometric figures, lines or points, the features can be stored in the form of coordinates or vectors with the smallest data size, thereby reducing the amount of data required when providing the feature map to a vehicle or robot.

[0022] According to one embodiment, the feature map is stored as a digital map or as a map layer of a digital map. As a result, the feature map can be used particularly flexibly. In particular, existing maps can be updated with the feature map or implemented as digital maps with minimal memory requirements.

[0023] According to another embodiment, the received measurement data are present as a point cloud and are assigned to a grid consisting of a plurality of cells. Preferably, the measurement data of each cell are averaged to create the intensity map. The cells of the digital map can be pixels, groups of pixels or polygons, for example. By averaging, local inconsistencies and fluctuations in the measured values ​​can be compensated. The measured values ​​can be determined in particular by the beams reflected or scattered back and subsequently detected by the radar sensor and / or lidar sensor.

[0024] According to another embodiment, a height map is created from the received measurement data, wherein, to create the height map, a weighted average is determined from the measurement data of each cell and the adjacent cells. This allows additional information to be extracted from the determined measurement data and used when creating the feature map.

[0025] According to another embodiment, in order to determine the height of the extracted features, information is received from the created height map and stored in the feature map. In this case, the height map can be superimposed on the feature map, and the corresponding properties or information of the height map can be transferred to the feature map. For this purpose, for example, the height or the gradient of the intensity at the characteristic position of the individual features can be used. This process can preferably be automated, wherein each cell of the height map is compared with each cell of the feature map.

[0026] According to another embodiment, the extracted features are stored as universally accessible features in a feature map. According to an advantageous configuration, the features are extracted and stored as geometric shapes, lines, points and / or point clouds, etc. Thus, objects, markers and characteristic or prominent shapes can be extracted from the measurement data of the surroundings and used to perform positioning. In particular, multiple static features can also be determined in a dynamic surrounding environment and used for precise position determination. In this case, the determination of the features is essentially independent of the sensor, so that the marker map can be used universally.

[0027] According to another embodiment, the received measurement data are implemented as position data and stored in a position curve map. Preferably, in the case of a successful comparison of the feature, the position determined is stored as a new measurement value in the position curve map. The feature map can be used, for example, to determine the position of a vehicle or a robot. Here, each current position is determined along the route, for example at defined time intervals, and stored in a position curve map or a position map. With the help of the position curve map, the route or trajectory traveled can be described. If the feature is found again in the feature map, the position in the feature map can be assigned to the vehicle or sensor from which the measurement data was determined. This position will then be stored in the position curve map as its own measurement.

[0028] According to another embodiment, the measurement data are acquired by at least one sensor, which is different from at least one sensor for creating the feature map. The extracted features can preferably be present in an abstract form, so that they are generally readable or comparable. Features in this form can be present as coordinates, for example in the form of text. The features in the coordinates can have, in particular, a starting point, an end point, an intermediate point, a direction, a length, a height, etc. This information can be stored with a particularly low memory requirement and used for comparison. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In the following, a preferred embodiment of the present invention is explained in more detail based on a highly simplified schematic diagram. It is shown here:

[0030] Figure 1 : A schematic diagram for illustrating an arrangement of the method according to the present invention;

[0031] Figure 2 : A schematic diagram for illustrating a method for creating a digital map according to an embodiment;

[0032] Figure 3 : A schematic diagram for illustrating a method for performing positioning according to an embodiment;

[0033] Figure 4 : Schematic intensity map;

[0034] Figure 5 : Schematic height map; and

[0035] Figure 6 : Stereoscopic view of feature map. DETAILED DESCRIPTION

[0036] Figure 1 A schematic diagram of an arrangement 1 is shown for illustrating a method 2 , 4 according to the invention.

[0037] The arrangement 1 has two vehicles 6, 8. Alternatively or additionally, the arrangement 1 can have a robot and / or another vehicle. According to the illustrated embodiment, the first vehicle 6 is used to perform the method 2 for creating a digital map, in particular for labeling a map. The second vehicle 8 is schematically shown in order to illustrate the method 4 for performing positioning within a digital map.

[0038] The first vehicle 6 has a control device 10 which is connected in a data-conducting manner to a machine-readable memory 12 and to a sensor 14. The sensor 14 can be a lidar sensor 14, for example.

[0039] The first vehicle 6 can sample the surroundings U and generate measurement data by means of the lidar sensor 14. The acquired measurement data can then be received and evaluated by the control device 10. The characteristic map created by the control device 10 can be provided to other road users and vehicles 8 via a communication link 16. The characteristic map can be stored in a machine-readable storage medium 12.

[0040] The second vehicle 8 also has a control device 11. The control device 11 is connected to a machine-readable storage medium 13 and a sensor 15 in a data-conducting manner. According to one embodiment, the sensor 15 is a camera sensor 15 and is also capable of acquiring measurement data of the surroundings U and transmitting them to the control device 11. The control device 11 can extract features from the measurement data of the surroundings U and compare them with features in a feature map received from the control device 11 via the communication link 16.

[0041] Figure 2 A schematic diagram is shown for explaining a method for creating a digital map according to one exemplary embodiment.

[0042] In a first step 18, measurement data of surroundings U are ascertained during a measuring journey of first vehicle 6 and received by control device 10. According to the exemplary embodiment, surroundings U are sampled using lidar sensor 14.

[0043] Subsequently, in step 19 , a SLAM method is performed using the measurement data received during the measurement journey. The trajectory of first vehicle 6 is determined by the SLAM method.

[0044] The received measurement data are converted 20 into a coordinate system of the trajectory. Alternatively, the trajectory can be converted into a coordinate system of the measurement data. This common coordinate system can be, for example, a Cartesian coordinate system.

[0045] An intensity map 30 is created 21 from the converted measurement data. Figure 4Such an intensity map 30 is shown. In particular, the measurement data can be present as a grid map with a plurality of cells 31, 32. The cells 31, 32 can be designed, for example, as pixels or groups of pixels. Each cell 31, 32 can have a location assigned to it, for example a GPS coordinate, corresponding to a coordinate system.

[0046] The intensity of each cell 31, 32 is then calculated. For this purpose, a mean value is calculated for all measured values ​​within the individual cells 31, 32. The intensity map 30 is thus constructed 21 based on the calculated mean value.

[0047] Additionally a height map 40 is created 22. The height map 40 is created based on a weighted average and is Figure 5 The weighted average is calculated for the measurement value in each cell 31 and the measurement data in the corresponding adjacent cell 32 .

[0048] In a further step 23, features are extracted from the intensity map 30. This can be done, for example, by an automated pattern recognition algorithm or manually by a human operator. For example, the transition between light and dark areas in the intensity map 30 can be considered as a possible pattern. From the height map 40, a characteristic can be assigned to each feature.

[0049] The determined features are stored 24 in a feature map 60 corresponding to their positions in the intensity map 30. The feature map 60 is stored in Figure 6 . Here, an exemplary lidar scan is superimposed with a plurality of pairs of features 62, 64, 66. The features 62, 64, 66 are exemplarily configured as lane markings 62, lane boundaries 64, and other markings 66 on the surface. The feature map 60 can be stored, for example, in a machine-readable storage medium 12 and provided via a communication link 16.

[0050] Figure 3 A schematic diagram for explaining a method 4 for performing positioning according to an embodiment is shown. The method 4 is performed by a control device 11 of a second vehicle 8, for example.

[0051] In step 25 , measured data of the surroundings U are ascertained by sensors 15 and transmitted to control device 11 . In addition, characteristic map 60 is received by control device 11 via communication link 16 . This can be achieved by a position map locator implemented in control device 11 .

[0052] In this case, the measurement data can be determined continuously or at defined time intervals and received by the control device 11. Furthermore, the control device 11 can receive odometer data.

[0053] In a further step 26, features 62, 64, 66 are extracted from the received measurement data. In this case, features 62, 64, 66 are compared 27 with the received feature map 60. In this comparison, an attempt is made to find features 62, 64, 66 detected in the vehicle interior on the feature map 60. In this case, the measurement data determined by odometry can limit the search area within the feature map 60. Since the feature map 60 has abstract and therefore generally applicable features 62, 64, 66, the measurement data determined by means of the camera sensor 15 can also be used for localization.

[0054] If a match is found between the features ascertained on the vehicle side and features 62 , 64 , 66 in feature map 60 , the position of vehicle 8 can be corrected or updated 28 .

Claims

1. A method (2) for creating a digital map by means of a control device (10), wherein: - receiving (18) measurement data of the surroundings (U) during the measurement journey, - executing (19) a SLAM method for determining a trajectory of the measurement journey based on the received measurement data, - converting (20) the received measurement data into a coordinate system of the trajectory, - using the converted measurement data to create (21) an intensity map (30), - extracting (23) features (62, 64, 66) from the intensity map (30) and storing (24) them in a feature map (60), wherein the features (62, 64, 66) are extracted by an algorithm for pattern recognition, wherein the extracted features (62, 64, 66) are stored in the feature map (60) as generally ascertainable features (62, 64, 66), wherein the received measurement data are assigned to a plurality of individual A grid of cells (31, 32), wherein, in order to create the intensity map (30), an average value of the measurement data of each cell (31, 32) is calculated, wherein a height map (40) is created from the received measurement data, wherein, in order to create the height map (40), a weighted average value is calculated from the measurement data of each cell (31) and of adjacent cells (32), wherein, in order to determine the height of the extracted features (62, 64, 66), information is received from the created height map (40) and stored in the feature map (60), wherein the height map (40) is superimposed on the feature map (60) and an attribute of the height map (40) is transferred to the feature map (60), wherein the attribute is the slope of the intensity at the feature position of each feature.

2. The method according to claim 1, wherein: The feature map (60) is stored as the digital map or as a map layer of the digital map.

3. The method according to claim 1 or 2, wherein: The received measurement data is available as a point cloud.

4. The method according to claim 1 or 2, wherein: The features (62, 64, 66) are extracted and stored as geometric shapes, lines, points and / or point clouds.

5. A method (4) for performing positioning by means of a control device (11), wherein: - receiving (25) measurement data and a feature map (60) of an environment (U), wherein the feature map (60) is created using a method according to any one of claims 1 to 4, - identifying and extracting features in the received measurement data (26), - comparing (27) at least one extracted feature with features (62, 64, 66) stored in the feature map in order to determine the position.

6. The method according to claim 5, wherein: The received measurement data are implemented as position data and stored in a position map, wherein, in the event of a successful comparison of the features (62, 64, 66), the ascertained position is stored (28) as a new measured value in the position map.

7. The method according to claim 5 or 6, wherein: The measurement data are ascertained by at least one sensor (15) which is different from at least one sensor (14) used to create the characteristic map (60). 8 . A control device ( 10 , 11 ) which is configured to carry out at least one of the methods ( 2 , 4 ) according to claim 1 .

9. A machine-readable storage medium (12, 13) having a computer program stored thereon, the computer program comprising instructions which, when the computer program is executed by a computer or a control device (10, 11), cause the computer or the control device to execute at least one of the methods according to any one of claims 1 to 7.

Images