Investigation procedure for open space determination and motor vehicle

Abstract

The invention comprises a determination method (200) for determining free space by means of a determination system (10) for a motor vehicle (100), the determination method (200) comprising: - Capturing (202) an environment and generating (204) a point cloud by a distance sensor (20) of the detection system (10), - Dividing (206) the area around the motor vehicle (100) into angled cells by a logic device (30) of the detection system (10), - Capturing (208) a distance between two points of the point cloud, - Comparing (210) the measured distance with at least one dimension of the motor vehicle (100), - Determining (212) a free space around the motor vehicle (100) by the logic device (30) based on the point cloud in the angle cells and the comparison. Furthermore, the invention comprises a motor vehicle (100) with a detection system (10).

Description

[0001] The present invention relates to a method for determining free space using a detection system for a motor vehicle. The present invention further relates to a motor vehicle with a detection system.

[0002] For the development of increasingly autonomous vehicles, free space estimation and determination is a fundamental requirement for most advanced automated driving functions, such as lane planning, evasive maneuvers, and automatic parking. Accurate free space detection guarantees the safety of all road users in an autonomous driving scenario, as only paths within a defined free space area are considered. Most free space detection algorithms use measurements from distance sensors like radar or lidar and some kind of occupancy grid to represent the free space. These sensors typically generate point clouds. Storing high-resolution occupancy grid maps is challenging due to their high memory requirements.

[0003] It is therefore the object of the present invention to eliminate, or at least partially eliminate, the disadvantages described above. In particular, it is the object of the present invention to provide a method for determining clearance by means of a determination system for a motor vehicle, as well as a motor vehicle with such a determination system, which enable reliable clearance determination in a simple and cost-effective manner.

[0004] The foregoing problem is solved by a determination method for determining free space using a determination system for a motor vehicle with the features of claim 1 and by a motor vehicle with the features of claim 10. Further features and details of the invention will become apparent from the dependent claims, the description, and the drawings. Features and details described in connection with the determination method according to the invention naturally also apply in connection with the motor vehicle according to the invention, and vice versa, so that the disclosure of the individual aspects of the invention is always, or can always be, mutually referenced.

[0005] According to a first aspect of the invention, the problem is solved by a determination method for determining free space using a determination system for a motor vehicle. The determination method comprises: - Capturing an environment and generating a point cloud using a distance sensor of the detection system, - Division of the vehicle's surroundings into angular cells by a logic device of the detection system, - Determining the distance between two points in the point cloud, - Comparing the measured distance with at least one dimension of the motor vehicle, - Determining a free space around the motor vehicle by the logic device based on the point cloud in the angle cells and the comparison.

[0006] The basic idea of ​​the invention is to divide the detected surroundings of a motor vehicle into angular cells for determining the free space, for example, using a Cartesian coordinate system. The detected surroundings are captured as a point cloud by the distance sensor. The free space is determined, particularly with reference to the position of the distance sensor of the detection system, by the logic device of the detection system and based on the point cloud within the angular cells. The angular cells are preferably to be understood as the surroundings of the motor vehicle divided into angular segments. Preferably, the angular cells are divided uniformly, and in particular identically. The logic device is defined as a component of the detection system and / or the motor vehicle.

[0007] The determination method according to the invention detects at least one distance between at least two points and compares the detected distance with at least one dimension of the motor vehicle, for example, the width of the motor vehicle. The determination method according to the invention is particularly advantageous because the free space around the motor vehicle is determined by the logic device based on the point cloud in the angular cells and the comparison.

[0008] Such a structured investigation procedure is particularly advantageous because it enables a reliable determination of free space in a straightforward and cost-effective manner.

[0009] According to a preferred embodiment of the invention, a determination method may include classifying the distance as large enough or too small for at least one dimension by comparison. By comparison, it is determined, in a clear and exemplary manner, whether the motor vehicle can pass through the measured distance between two points. Preferably, the environment and situations of the motor vehicle are analyzed for possible routes, and only those routes that are sufficiently wide for the motor vehicle are identified.

[0010] According to a preferred further development of the invention, a determination procedure may be provided to further include: - Generating virtual points in unoccupied angle cells by the logic device, in particular wherein the generation is based on comparison and / or wherein the virtual points are generated between two points of the point cloud.

[0011] Generating and / or inserting virtual points into unoccupied azimuthal cells is particularly advantageous, for example, to fill a gap between two points when the vehicle, as described in the previous section, cannot pass through a certain distance and / or gap. In practice, the problem often arises that with sparse point clouds, such as those generated by radar sensors, not every azimuthal cell is occupied. If the adjacent azimuthal cells are occupied and the vehicle would not fit through them anyway due to its size, it is advantageous for the detection process to fill these unoccupied azimuthal cells with virtual points. The problem is solved, as described clearly and by way of example, by inserting virtual points for unoccupied azimuthal cells to fill the gap, provided that this gap is small enough for the vehicle to pass through.The gap measurement is the arc distance between two adjacent real points. The azimuth angle for each virtual point is set, for example, to the center of the respective angle cell, and the range of each virtual point is set to the minimum range of one of the two adjacent real points in the point cloud. This restricts these areas to make the free space determination more conservative and reliable, and to reduce or prevent the formation of sharp peaks and / or outliers at the free space boundary.

[0012] According to a preferred further development of the invention, a determination procedure may be provided to further include: - Receiving data from at least one other sensor device of the detection system and / or the motor vehicle, whereby the determination of the free space is additionally based on the received data.

[0013] Preferably, at least one additional sensor device can provide data about the vehicle's surroundings to enable more reliable and / or detailed determination of the free space. This additional sensor device is defined as a component of the detection system and / or the vehicle. The detection system preferably has a data interface for sending and / or receiving data. The additional sensor device is preferably configured as an environmental sensor, an optical sensor, and / or a camera. For example, the additional sensor device enables the detection and provision of environmental data. One example of this is the projection of detected radar points into the bounding boxes of an additional camera detection system and the performance of data associations to assign object designations to the points.

[0014] According to a preferred further development of the invention, a determination procedure may be provided to further include: - Classifying and / or grouping points of the point cloud and / or the virtual points, especially based on the received data.

[0015] Classification and / or grouping is preferably performed by the logic unit of the detection system. For example, and preferably, point labels are added to the individual points of the point cloud based on data from other sensors and / or a camera. These point labels are used to recognize classifications and / or groupings of points, and virtual points are inserted into the angular cells of the grouping, particularly when the vehicle can pass through the gap. Classification and / or grouping of points is preferably understood as the logical and / or spatial connection of points in the point cloud to an object, such as another motor vehicle, an environmental object, and / or a road traffic object.

[0016] According to a preferred further development of the invention, a determination procedure may be provided to further include: - Generating virtual points in angle cells by the logic device, in particular wherein the generation is based on classification and / or grouping and / or on the received data.

[0017] Preferably, two adjacent points, for example, two spaced-apart tires, are detected by the distance sensor and, with the help of additional data from a camera, assigned to a common object, such as a truck. Based on radar data alone, the distance between the tires would be large enough for the vehicle to pass through, because the radar device, for example, sees between the detected points underneath the truck. The additional data from the example camera allows the detection method to link the spaced-apart points to a single object and, for example, to generate virtual points between the tires to support clearance determination and to define the clearance between the tires.

[0018] According to a preferred further development of the invention, a determination procedure may be provided to further include: - Determining the direction of movement of the motor vehicle by the detection system, whereby the determination of the free space is additionally carried out on the basis of the determined direction of movement.

[0019] Determining the direction of movement of the motor vehicle is an advantageous addition to the investigation procedure. The direction of movement of the motor vehicle is to be understood as a linear direction of movement, as a movement pattern, and / or as a movement area. Preferably, the clearance is determined taking the direction of movement into account. By way of example, angular cells that the motor vehicle can reach in the next few movements or within a certain time period are primarily or exclusively considered for determining the clearance.

[0020] According to a preferred further development of the invention, a determination procedure may be provided to further include: - Discretizing the point cloud into the angle cells by the logic device and / or - Discretizing the virtual points into the angle cells by the logic device.

[0021] Discretization is a fundamental concept in mathematics and refers to all methods that transform a continuous problem into a discrete one. In the present case of functions, this preferably means that the captured points of the point cloud are approximated by values ​​at a finite number of support points or grids. To illustrate this, the points closest to the vehicle form the coordinates of the free space boundary for each azimuth cell (0 ... 360 degrees), while preferably all other points are deleted. Preferably, the points of the point cloud from the distance sensor and / or the virtual points are discretized into the angle cells.

[0022] According to a preferred embodiment of the invention, a detection method may include discretizing the point cloud and / or the virtual points, as well as discretizing the angles of the points and / or the depth of the points within the angle cells. The angle and / or depth of the points is preferably understood in relation to the distance sensor and / or the position of the distance sensor. Such discretization preferably leads to a very effective representation of the free space without significant loss of quality in free space detection.

[0023] According to a second aspect of the invention, the problem is solved by a motor vehicle comprising at least one determination system for determining free space. The determination system is designed to carry out the determination method according to the first aspect. The described motor vehicle offers all the advantages already described for the determination method according to the first aspect of the invention.

[0024] A determination method according to the invention and a motor vehicle are explained in more detail below with reference to the drawings. The drawings schematically show: Fig. 1. In a perspective view, a motor vehicle with a system for determining free space, Fig. 2. In a perspective view, a system for determining open space, and Fig. 3 in a flowchart an embodiment of the investigation method according to the invention.

[0025] Elements with the same function and mode of operation are in the Fig. Items 1 to 3 are each labelled with the same reference numerals.

[0026] In Fig. Figure 1 schematically shows a perspective view of a motor vehicle 100 with a free space determination system 10. The determination system 10 comprises a distance sensor 20 for detecting 202 (not shown) the surroundings and generating 204 (not shown) a point cloud. The determination system 10 includes a logic device 30 for dividing 206 (not shown) the surroundings of the motor vehicle 100 into angular cells. The logic device 30 is further configured to detect 208 (not shown) the distance between two points in the point cloud, to compare 210 (not shown) the detected distance with at least one dimension of the motor vehicle 100, and to determine 212 (not shown) a free space around the motor vehicle 100 based on the point cloud in the angular cells and the comparison.

[0027] In Fig. Figure 2 schematically shows a determination system 10 for determining free space in a perspective view. The determination system 10 comprises a distance sensor 20 and a logic device 30. Fig. Figure 2 schematically shows the division 206 of the area around the motor vehicle 100 into angled cells and the determined free space by the determination method 200 according to the invention (not shown).

[0028] In Fig. Figure 3 schematically shows a flowchart illustrating an embodiment of the determination method 200 according to the invention. For improved clarity, in Fig.3 Only the reference numerals of the process steps are given. The investigation process 200 comprises, in a first process step, the acquisition 202 of an environment and the generation 204 of a point cloud by a distance sensor 20 of the investigation system 10. The investigation process 200 comprises, in a further process step, the division 206 of the environment of the motor vehicle 100 into angular cells by a logic device 30 of the investigation system 10. The investigation process 200 comprises, in a further process step, the acquisition 208 of a distance between two points of the point cloud. The investigation process 200 comprises, in a further process step, the comparison 210 of the acquired distance with at least one dimension of the motor vehicle 100. The investigation process 200 comprises, in a further process step, the determination 212 of a free space around the motor vehicle 100 by the logic device 30 based on the point cloud in the angular cells and the comparison.The investigation procedure 200 comprises, in a further procedural step, the generation 214 of virtual points in unoccupied angle cells by the logic device 30. The investigation procedure 200 comprises, in a further procedural step, the reception 216 of data from at least one further sensor device of the investigation system 10 and / or the motor vehicle 100, wherein the determination 212 of the free space is additionally carried out on the basis of the received data. The investigation procedure 200 comprises, in a further procedural step, the classification and / or grouping 218 of points of the point cloud and / or the virtual points. The investigation procedure 200 comprises, in a further procedural step, the generation 220 of virtual points in angle cells by the logic device 30.The investigation procedure 200 comprises, in a further procedural step, the determination 222 of a direction of movement of the motor vehicle 100 by the investigation system 10, whereby the determination 212 of the free space is additionally carried out on the basis of the determined direction of movement. The investigation procedure 200 comprises, in a further procedural step, the discretization 224 of the point cloud into the angular cells by the logic device 30. The investigation procedure 200 comprises, in a further procedural step, the discretization 226 of the virtual points into the angular cells by the logic device 30.

Claims

[1] Investigation procedure (200) for determining free space by means of an investigation system (10) for a motor vehicle (100), comprising the investigation procedure (200): - Capturing (202) an environment and generating (204) a point cloud by a distance sensor (20) of the detection system (10), - Dividing (206) the area around the motor vehicle (100) into angled cells by a logic device (30) of the detection system (10), - Capturing (208) a distance between two points of the point cloud, - Comparing (210) the measured distance with at least one dimension of the motor vehicle (100), - Determining (212) a free space around the motor vehicle (100) by the logic device (30) based on the point cloud in the angle cells and the comparison. [2] Investigation procedure (200) according to claim 1, characterized by, that by comparing (210) the distance is classified as large enough or too small for at least one size measure. [3] Investigation procedure (200) according to one of the preceding claims, characterized by , that the investigation (200) further includes: - Generating (214) virtual points in unoccupied angle cells by the logic device (30), in particular wherein the generation (214) is based on the comparison (210) and / or wherein the virtual points are generated between two points of the point cloud. [4] Investigation procedure (200) according to one of the preceding claims, characterized by , that the investigation (200) further includes: - Receiving (216) data from at least one further sensor device of the detection system (10) and / or the motor vehicle (100), wherein the determination (212) of the free space is additionally carried out on the basis of the received data. [5] Investigation procedure (200) according to one of the preceding claims, characterized by , that the investigation (200) further includes: - Classifying and / or grouping (218) points of the point cloud and / or the virtual points, in particular based on the received data. [6] Investigation procedure (200) according to one of the preceding claims, characterized by , that the investigation (200) further includes: - Generating (220) virtual points in angle cells by the logic device (30), in particular wherein the generation (220) is based on classifying and / or grouping (218) and / or on the received data. [7] Investigation procedure (200) according to one of the preceding claims, characterized by , that the investigation (200) further includes: - Determining (222) a direction of movement of the motor vehicle (100) by the determination system (10), wherein the determination (212) of the free space is additionally carried out on the basis of the determined direction of movement. [8] Investigation procedure (200) according to one of the preceding claims, characterized by , that the investigation (200) further includes: - Discretizing (224) the point cloud into the angle cells by the logic device (30) and / or - Discretizing (226) the virtual points into the angle cells by the logic device (30). [9] Investigation procedure (200) according to claim 8, characterized by , that discretizing (224) the point cloud and / or discretizing (226) the virtual points includes discretizing the angles of the points and / or the depth of the points in the angle cells. [10] Motor vehicle (100) comprising at least one detection system (10) for determining free space, characterized by, that the investigation system (10) is designed to carry out the investigation procedure (200) according to one of the preceding claims.

Images