Efficient provision of occupancy information for the surroundings of a vehicle

Abstract

A method for efficiently providing occupancy information for the surroundings of a vehicle, comprising: Receiving sensor measurements of the vehicle’s surroundings; Determining the occupancy of the environment by obstacles based on the sensor measurements; wherein occupancies in a first section of the environment are indicated in a first coordinate system, namely a polar coordinate system, by an angle indication and a distance indication; wherein occupancies in a second section of the environment are specified in a second coordinate system, in particular by two values, wherein the second coordinate system differs from the polar coordinate system; wherein the second coordinate system is a Cartesian coordinate system; and characterized in that the first value is based on the distance of the projection of the occupancy onto a trajectory, measured along the trajectory, from the vehicle, the trajectory is the path traversed by the vehicle; the projection is perpendicular to the trajectory.

Description

[0001] The invention relates to a method for efficiently providing occupancy information for the surroundings of a vehicle and a correspondingly configured computing device.

[0002] In the future, motor vehicles will have a multitude of driver assistance systems that warn the driver of collisions and, if necessary, attempt to avoid collisions through intervention. Examples of such driver assistance systems include emergency braking, lane keeping assist, blind spot assist, parking assist, and so-called automatic cruise control assist (ACC), particularly for motorway driving. In order to provide these functions, knowledge of the vehicle's surroundings is crucial for driver assistance systems. For this purpose, the surroundings are scanned or recorded using one or more sensors such as radar, lidar, cameras, ultrasonic sensors, or similar state-of-the-art sensors. Using signal processing methods that are also state-of-the-art, the occupancy of an obstacle in the surroundings can then be detected.The occupancy indicates that the area surrounding this zone cannot be accessed by the vehicle.

[0003] It is known to date that the environment is divided into uniform, preferably rectangular, areas, and then the occupied areas are determined. This creates an occupancy grid. Publication WO 2013 / 060 323 A1 describes such an occupancy grid. In such a system, the size of the areas is often specified to correspond to the highest requirements of the driver assistance systems. This division of the environment generates very large amounts of data. At the same time, reducing the size of the areas to reduce data can provide insufficient accuracy.

[0004] Similar systems for describing the environment are known from the publications US 2006 / 0 125 680 A1; BOREN-STEIN, J. ; KOREN, Y.: The vector field histogram - fast obstacle avoidance for mobile robots. In: IEEE Transactions on Robotics and Automation, Volume 7, No. 3, pp. 278-288, 1991. - Electronic ISSN: 2374-958X; US 5 006 988 A and US 2008 / 0009 968 A1.

[0005] The object underlying the invention is to efficiently provide occupancy information for the surroundings of a vehicle.

[0006] The problem is solved by the method and the computing device according to the independent claims. Advantageous further developments are defined in the dependent claims.

[0007] In one aspect, a method for efficiently providing occupancy information for the surroundings of a vehicle comprises: receiving sensor measurements of the surroundings of the vehicle; determining the occupancy of the surroundings by obstacles based on the sensor measurements; wherein occupancies in a first section of the surroundings are each specified by an angle specification and a distance specification in a first coordinate system, namely a polar coordinate system; wherein occupancies in a second section of the surroundings are each specified by two value specifications in a second coordinate system, wherein the second coordinate system differs from the polar coordinate system.

[0008] By using different coordinate systems, a polar coordinate system on the one hand and a Cartesian coordinate system on the other, for example, it is possible to use the coordinate system that offers advantageous properties for those sections of the environment. These properties include, for example, the storage space required compared to the relevant information. For example, in the section of the environment in front of a vehicle, a description of detected obstacles in polar coordinate form is useful: the decision as to which direction to drive is reflected in the angle of the polar coordinate system. Furthermore, with a polar coordinate system, the representation near the vehicle is more accurate. The position of nearby obstacles is recorded more accurately using a polar coordinate system.At the same time, the polar coordinate system is not a suitable representation for reversing when parking or for observing traffic behind. In this case, a Cartesian representation, or a curvilinear representation for the rear area of ​​the vehicle, is suitable. For example, an obstacle in the first section of the surrounding area is described by specifying the angle and distance from the vehicle. Similarly, an obstacle in the second section of the surrounding area can be described in x- and y-coordinates of a Cartesian coordinate system.

[0009] In order to reduce the storage and computing requirements for describing the environment, the angular information of the polar coordinate system can be discretized. Each angular information then represents an angular range. When determining the occupancy, it is then determined whether an obstacle is present in the angular range and at what distance. The inaccuracies resulting from a given discretization in a polar coordinate system for the area in front of a vehicle are less serious than the inaccuracies that must be accepted for a comparable reduction in storage requirements in a Cartesian coordinate system due to the correspondence of the angular information to the selection of the direction of travel. The use of two forms of representation of the environment thus enables a discretization whose inaccuracies are adapted to the characteristics of a vehicle's locomotion and the requirements of driver assistance systems and have as little influence as possible.

[0010] An obstacle represents a boundary of the vehicle's free space. Each angular range can be considered a segment. The size of the angular ranges can be selected according to the hardware's capabilities.

[0011] In a further development, determining occupancy includes: determining a reliably detected obstacle; and determining the obstacle closest to the vehicle; whereby for each angular range, only the reliably detected obstacle and the nearest detected obstacle are specified as detected obstacles, provided that it is present in the angular range. When determining the obstacles, a probability can be determined for each presence. A reliably detected obstacle is one whose determined probability exceeds a threshold value, or whose probability is high compared to the probabilities of other obstacles. In this way, the amount of data to be processed and stored can be limited to relevant information. The information is compressed. For many assistance systems, the nearest and most reliably detected obstacles represent the most important decision-making basis.For example, in a driver assistance system that automatically initiates an evasive maneuver, even the uncertain presence of an upcoming obstacle can prevent intervention in that direction. At the same time, interventions are only triggered if obstacles are reliably detected. In other words, the detection of the upcoming obstacle is optimized to prevent an action from being triggered (obstacles must not be overlooked), whereas the reliable detection of obstacles is optimized to trigger an action (obstacles must be detected with a high degree of probability).

[0012] In one embodiment, the second coordinate system is a Cartesian coordinate system, and the first value represents a region of the environment. For example, the Cartesian coordinate system can include values ​​in the x- and y-directions. The first value can then specify ranges of e.g. 10 m each, so that one x-value covers the range from 0 m to 10 m, the next x-value covers the range from 10 m to 20 m, and so on. A distance range can be considered a segment.

[0013] In a preferred embodiment, the first value is based on the distance of the projection of the respective occupancy onto a trajectory, measured along the trajectory, from the vehicle; where the trajectory is the path traveled by the vehicle; where the projection is perpendicular to the trajectory. Instead of a Cartesian coordinate system, a curved variant of the Cartesian coordinate system is used, with the x-axis of the Cartesian coordinate system lying on the path traveled by the vehicle. The x-value of an obstacle is then calculated by a (point-by-point) perpendicular projection of the obstacle onto the x-axis. Such a coordinate system is sometimes called curvilinear. In the rear area, the curvilinear representation (related to the ego trajectory) is most suitable, since the past route is known and the right and left sides of the route can thus be clearly distinguished.Furthermore, this processing is particularly suitable for functions based on it, such as lane change assist, etc. For small detection areas behind the vehicle, a linear representation can be used instead of the curvilinear representation. The aforementioned advantages also apply to the Cartesian representation for straight stretches and, to a lesser extent, for curves.

[0014] Even in a curvilinear coordinate system, value ranges can be used, and only reliably detected and nearest obstacles can be displayed to reduce data volumes. This results in the advantages described above for the polar coordinate system.

[0015] In an advantageous further development, the occupancies in a third section of the environment are specified using a third coordinate system. A further coordinate system enables further compression of the data volumes, tailored to the requirements of the driver assistance systems.

[0016] In an advantageous development, the angle data of the first coordinate system each represent an angular range; the angle data of the third coordinate system each represent an angular range that is larger than the angular ranges of the first coordinate system. The first section can comprise the extension of the vehicle's longitudinal axis, with the third section adjacent to the first section. In this way, the section directly in front of the vehicle is recorded with greater accuracy than the section that is essentially almost to the side of the vehicle. This gradation of accuracy also serves to reduce data, which has little impact on the functioning of driver assistance systems.

[0017] In a further development, the method further includes: linking neighboring identified occupancies, in particular linking neighboring detected obstacles, to form a continuous occupancy or a continuous obstacle. The link can be conceptualized as a polyline. This allows for the description of frequently encountered boundaries such as guardrails or house walls.

[0018] In another aspect, a computing device is configured to perform one of the above methods. The computing device may be a computer with a stored computer program or an application-specific circuit. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 shows a schematic division of the environment into segments. Fig. 2 schematically shows a further division of the environment into segments according to an embodiment. Fig. Figure 3 shows a schematic variant of the detection of occupancy in the environment. Fig. Figure 4 schematically shows another variant of the detection of occupancy in the environment.

[0019] Like reference numerals refer to corresponding elements throughout the figures. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] Fig. Figure 1 schematically shows a division of the surroundings of a vehicle 1 into segments. Vehicle 1 has sensors for environment detection and a computing device for creating an environment map that indicates occupancy. To divide the environment, the environment is divided into four sections 2, 3a, 3b, and 4. In sections 2, 3a, and 3b, which are essentially located in front of vehicle 1, a polar coordinate system is used to indicate occupancy. In section 4, which is essentially located behind vehicle 1, the occupancy is indicated using a Cartesian coordinate system. In each of the sections, regions are formed for discretization. In sections 2, 3a, and 3b, angular regions are formed, each represented by an angle specification. In sections 3a and 3b, each angular region is 20°. In section 2, each angular region is 10°. In section 2, the division is therefore finer.In Section 4, the distance in the x-direction is also divided into distance ranges. The individual distance ranges and angle ranges form segments.

[0021] Using the sensor measurements of vehicle 1, obstacles in the surroundings of vehicle 1 are detected. The result of this detection is a statement about the location of an obstacle and the probability of the detection. However, to specify the occupancy (for use by assistance systems, for example), only the next detected obstacle (i.e., the next detected occupancy), regardless of the probability with which it was detected, and the detected obstacle that was detected with the highest probability in this segment are specified. This reduces the memory requirements of the resulting environment map due to a selection of information to be specified, which is oriented towards the requirements of the driver assistance systems. Fig. 1, the nearest obstacles 6 are represented by dashed circles. The most reliably detected obstacles 5 are represented by circles with a solid line. If the most reliably detected obstacle is also the nearest, only that one obstacle is indicated in this segment.

[0022] Fig. Figure 2 shows a schematic division of the environment into segments according to an embodiment. This example builds on that of Fig. 1, where in section 4a, the information is given not in a Cartesian coordinate system, but in a curvilinear coordinate system. The trajectory forming the x-line corresponds to the path the vehicle traveled. Section 4a is also divided into distance ranges, which run perpendicular to the trajectory.

[0023] Fig. Figure 3 schematically shows a variant of the detection of occupancy in the surrounding area. The computing device detects and indicates free space boundaries. Even with these continuous free space boundaries, a distinction is made between the nearest detected boundary 8 and the most reliably detected boundary 7. Only these two boundaries are indicated to adapt processing to the performance of the computing systems.

[0024] Furthermore, attributes of detected obstacles or occupancies can also be stored, such as movable, immovable, type classifier (vehicle, pedestrian, ...), speeds, etc.

[0025] Fig. Figure 4 schematically shows another variant of detecting occupancy in the surrounding area. In this further variant, individually detected neighboring obstacles are linked to form a polyline. Such a line can, for example, well describe road boundaries or house walls.

Claims

[1] A method for efficiently providing occupancy information for the surroundings of a vehicle, comprising: Receiving sensor measurements of the vehicle’s surroundings; Determining the occupancy of the environment by obstacles based on the sensor measurements; wherein occupancies in a first section of the environment are indicated in a first coordinate system, namely a polar coordinate system, by an angle indication and a distance indication; wherein occupancies in a second section of the environment are specified in a second coordinate system, in particular by two values, wherein the second coordinate system differs from the polar coordinate system; where the second coordinate system is a Cartesian coordinate system; and characterized by , that the first value is based on the distance of the projection of the occupancy onto a trajectory, measured along the trajectory, from the vehicle, the trajectory is the path traversed by the vehicle; the projection is perpendicular to the trajectory. [2] Method according to claim 1, wherein the angle information each represents an angular range; wherein, when determining the occupancy, it is determined whether an obstacle is present in the respective angular range and at what distance. [3] The method of claim 2, wherein determining the occupancy comprises: Determining a reliably detected obstacle; and Determining the obstacle closest to the vehicle; For each angle specification, only the reliably detected obstacle and the next detected obstacle are specified as detected obstacles, provided they exist in the corresponding angle range. [4] Method according to one of the preceding claims, wherein the first section covers a section of the surroundings in the direction of travel of the vehicle in front of the vehicle. [5] Method according to one of the preceding claims, wherein the second coordinate system covers a portion of the surroundings in the direction of travel of the vehicle behind the vehicle. [6] Method according to one of the preceding claims, wherein the first value represents a distance range. [7] The method of claim 6, wherein determining the occupancy comprises: Determining a reliably detected obstacle; and Determine the next obstacle; For each distance indication, only the reliably identified obstacle and the next identified obstacle are indicated as identified obstacles, provided that they are present in the corresponding distance range. [8] Method according to one of the preceding claims, wherein occupancies in a third section of the environment are indicated by a third coordinate system. [9] Method according to claim 8, where the angle values ​​of the first coordinate system each represent an angular range; where the angle values ​​of the third coordinate system each represent an angular range that is larger than the angular ranges of the first coordinate system. [10] A method according to claim 8 or 9, wherein the first section comprises the extension of the longitudinal axis of the vehicle and wherein the third section is adjacent to the first section. [11] Method according to one of the preceding claims, further comprising: Linking neighboring determined occupancies, in particular linking neighboring detected obstacles, to form a continuous occupancy or a continuous obstacle. [12] Computing device, wherein the computing device is configured to carry out a method according to any one of the preceding claims.

Images