Method for generating surroundings information that describes the surroundings of a vehicle, surroundings detection device, vehicle, and computer program product

Abstract

The invention relates to a method for generating surroundings information that describes the surroundings of a vehicle (1), the vehicle (1) comprising a sensor system having at least one surroundings sensor (4), and the surroundings information comprising at least one grid map, the grid map having a plurality of cells, each of which is assigned to a sub-region of the surrounding region of the vehicle and each of which is assigned an occupancy value, said method comprising the steps of: a) emitting a sensor signal using the surroundings sensor (4); b) receiving a reflected signal component of the sensor signal using the surroundings sensor (4); c) assigning the reflected signal component to one or more cells of the grid map based on a current position of the surroundings sensor (4); d) changing the occupancy values of each of the cells to which the reflected signal component is assigned, in each case by a variable count value, the count value depending on an angle between the cell in question and the current position of the surroundings sensor (4) and / or depending on a detection intensity assigned to the cell in question.

Description

[0001] Description

[0002] Method for generating environmental information describing an environment of a vehicle, environmental detection device, vehicle and computer program product

[0003] The invention relates to a method for generating environmental information describing the surroundings of a vehicle, wherein the vehicle has a sensor device with at least one environmental sensor, and the environmental information comprises at least one raster map. The raster map has a plurality of cells, each of which is assigned to a sub-area of ​​the surrounding area of ​​the vehicle and to which an occupancy value is assigned. Furthermore, the invention relates to an environmental detection device, a vehicle, and a computer program product.

[0004] An environmental sensor in a motor vehicle is typically used for environmental detection. For example, a distance, or in other words, the distance to an object, is determined based on the propagation time between a transmitted measurement signal and a received echo of the measurement signal reflected from an object. Such an environmental sensor is usually designed as a radar sensor or an ultrasonic sensor, with ultrasonic sensors being particularly widespread and installed in almost all passenger vehicles today.

[0005] Ultrasonic sensors typically comprise a transmitter that emits ultrasonic signals as measurement signals propagating through the air at the speed of sound of approximately 340 meters per second. For this purpose, a membrane of the ultrasonic sensor is typically excited to mechanical vibrations using a corresponding transducer element. The ultrasonic signal is reflected by surrounding objects as an echo and detected by a receiver of the ultrasonic sensor. Based on the time difference between the time of transmission and the time of reception, the distance to the object can be determined, taking into account the propagation speed of the ultrasonic signal. The amplitude of the reflected ultrasonic signal or echo can also be determined.

[0006] Ultrasonic sensors are typically used in motor vehicles to detect the surroundings within a range of up to approximately seven meters. Ultrasonic sensors are particularly important in semi-automatic or automatic driving maneuvers, especially in connection with parking applications, such as parking distance measurement, parking space searches, or parking. The vehicle is typically moved relative to the objects, with a measurement cycle being performed at predetermined times during the movement. During each measurement cycle, an ultrasonic sensor emits an ultrasonic signal.

[0007] Methods and corresponding assistance systems are already known from the state of the art which, with the aid of ultrasonic sensors, provide the driver with various information about the surroundings of the motor vehicle and assist him in maneuvering the motor vehicle and, in particular, in locating a parking space and parking the motor vehicle in the parking space.

[0008] For example, there are assistance systems equipped with parking space location that indicate to the driver whether a parking space is available in the immediate vicinity of the vehicle or whether an existing parking space is large enough for the vehicle to park in. To reliably locate and measure a parking space, such assistance systems require information about objects in the vicinity of the vehicle, such as parked vehicles, curbs, walls, or barriers.

[0009] Such objects can be detected, for example, by assigning reflected ultrasonic signals, which allow distance determination, to different sections or cells of an environmental map. Depending on how often a reflected ultrasonic signal has already been assigned to a section or cell, the presence of an object in that section or cell can be determined.

[0010] DE 10 2020 215253 A1 describes a method for detecting parking spaces using at least one ultrasonic sensor of a vehicle. A reflected signal component is assigned to a set of multiple cells of a grid of an environment map based on the propagation time of the ultrasonic signal between the transmission and reception of the reflected signal component, and the occupancy values ​​of these cells are each increased by a specific value. Occupancy information is then determined based on the occupancy values ​​of the cells in the set of cells by determining the cell in the set of cells that has the maximum occupancy value. If the maximum occupancy value exceeds a first threshold, a counter assigned to the cell with the maximum occupancy value is changed by an incremental value.This counter is then compared with a second threshold and an occupancy status of the cell is determined depending on the comparison result.

[0011] However, using such environment maps can pose challenges in distinguishing differently shaped objects.

[0012] US 2003 / 0128153 A1 discloses a method in which measured values ​​are obtained using distance sensors. The measured values ​​are used to create a list of potential objects, each of which is assigned a plurality of measured values. For each object in the list, a plurality of different positions are then determined, which the object can have assuming various possible object shapes. Depending on the difference between the positions determined for the potential objects and the measured distances, each object shape is assigned an error measure, with the object shape with the smallest error measure being assumed to be the actual object shape.The invention is based on the object of specifying an improved method for generating environmental information describing an environment of a vehicle, which method in particular reduces the occurrence of false positive and / or false negative object assignments to the cells of environmental information determined as a raster map.

[0013] To achieve this object, the invention provides for a method of the type mentioned at the outset to comprise the following steps: a) transmission of a sensor signal by the environmental sensor; b) reception of a reflected signal component of the sensor signal by the environmental sensor; c) assignment of the reflected signal component to one or more cells of the raster map starting from a current position of the environmental sensor; d) changing the respective occupancy values ​​of the cells to which the reflected signal component is assigned by a variable count value, wherein the count value depends on an angle between the respective cell and the current position of the environmental sensor and / or on a detection intensity assigned to the respective cell.

[0014] First, a sensor signal is emitted by the vehicle's environmental sensor. The environmental sensor can, in particular, be an ultrasonic sensor, which emits an ultrasonic signal into the vehicle's surroundings as a sensor signal. Alternatively, other embodiments of the environmental sensor are also possible, for example, as a radar sensor, a lidar sensor, or similar. When emitting the sensor signal, only a single environmental sensor can transmit a sensor signal, or several environmental sensors can each transmit sensor signals simultaneously or sequentially. Subsequently, a reflected signal component of the sensor signal is received by the environmental sensor. The reflected signal component (hereinafter also referred to as the echo) can be received either by the same environmental sensor that transmitted the sensor signal or by another environmental sensor (so-called cross-echo).

[0015] Furthermore, a raster map is provided that relates to an area surrounding the vehicle. The raster map can, for example, be provided or created as a data structure in a data memory of a control device configured to carry out the method. The raster map has a plurality of cells, each cell being assigned to a sub-area of ​​the area surrounding the vehicle. The raster map or the cells of the raster map are preferably moved along with the respective position of the vehicle, i.e., the raster map has a fixed reference to the vehicle position and does not relate to a stationary area surrounding it.

[0016] Each cell of the raster map is assigned an occupancy value. The occupancy value of a cell is a measure of how often received echoes have already been assigned to that cell. The occupancy value can be used, in particular, to determine whether the respective cell is occupied, i.e., whether an object is located within it. Depending on their size, objects can also extend across multiple cells. The raster map can be viewed as a probabilistic map.

[0017] A reflected signal component received by at least one of the environmental sensors is assigned to one or more of the cells. The assignment can be made in particular based on a distance value determined from the received signal component. The reflected signal component can be assigned to all cells that are at least partially within the determined distance. The assignment of the signal component to the cells then causes the occupancy values ​​of these cells to be changed due to the reception of the signal component. For cells to which the signal component has not been assigned, the occupancy value is correspondingly not changed. The assignment of the reflected signal component takes into account the current position of the environmental sensor, i.e. it is made in particular as a function of the determined distance value in relation to the current sensor position or in relation to the current position of the vehicle.

[0018] The respective occupancy values ​​of the cells to which the reflected signal portion is assigned are each changed by a variable count. The count by which the occupancy value is changed depends on an angle between the respective cell and the current position of the environmental sensor and / or on a detection intensity assigned to the respective cell. When changing the occupancy value of a cell, the angle of this cell to the current sensor position and / or the detection intensity, in particular the detection frequency and / or detection duration, that has been assigned to the cell so far are taken into account. This means that when the reflected signal portion is assigned to several cells, the respective occupancy value of the individual cells is at least partially changed, in particular increased, by a different count.

[0019] By taking into account the angle between the respective cell and the current sensor position and / or the detection intensity, the occurrence of false positive object detections when determining the environmental information can be advantageously reduced.

[0020] Changing the occupancy value depending on the angle makes it possible for point-shaped objects and linear objects to have more similar occupancy values ​​when determining object occupancy than would be the case without taking the angle into account or when using a fixed or non-angle-dependent count value.

[0021] If a point-like object is in the vicinity of the moving

[0022] vehicle, different echoes are always assigned to at least the cell located at the actual object position based on the distance values ​​described by the echoes. In other words, the circles or circle segments described by the distance values ​​assigned to the reflected signal components all intersect at the same point, so that when the reflected signal component is assigned to all cells intersected by the circles or circle segments, a change in the occupancy value occurs for each reflected signal component in at least one cell. This relationship is illustrated graphically in Fig. 8.

[0023] In the case of a linear object, for example, another vehicle parked parallel to the passing vehicle, the echoes are always reflected from the same direction and at the same distance, or at the same distance value. This leads to cells between the linear object and the ambient sensor experiencing a more frequent change in their occupancy value. This relationship is illustrated graphically in Fig. 9.

[0024] The aforementioned differences mean that, in the case of a point-shaped object, the reflected signal component is very frequently assigned to a single cell, so that the occupancy value is also very frequently changed, especially increased. This is not the case with a linear object, as the reflected signal components are each assigned to different cells, which would accordingly each have a lower total occupancy value if a constant count value were used to increase the occupancy value.

[0025] When comparing the occupancy value with a threshold value for distinguishing whether a cell is occupied or not, point-like objects would be displayed in particular, while linear objects could be overlooked if the threshold value is set too high. Conversely, using a threshold value that is too low could result in too many false positive object detections between the object and the environmental sensor. By taking the angle into account through the use of the angle-dependent count value, this effect can be at least partially compensated for during detection, thereby improving the raster map and thus the environmental information.

[0026] Particularly when using the method according to the invention in a motor vehicle, the problem arises that the motor vehicle is often moved at highly variable speeds. For example, before performing a parking maneuver, which is to be carried out based on the determined environmental information, the vehicle can reach very low speeds, even temporarily coming to a standstill.

[0027] This means that objects located in the immediate vicinity of the motor vehicle are detected more often than objects further away. In other words, objects closer to the vehicle have a higher detection intensity, in particular a longer detection duration and / or a higher detection frequency, than cells further away. As a result, cells closer to the vehicle experience a change in their occupancy value more often than cells further away. So, for example, when using a constant count to increase the occupancy value, closer objects appear clearer, or when using a threshold value, analogous to the previous description, this can lead to individual objects being lost or to an excessive number of false positive object detections.By taking into account a count value that depends on the detection intensity, this effect can be taken into account and at least partially compensated, so that the raster map or the environmental information can also be improved.

[0028] It is possible to determine several different occupancy information items in parallel, particularly based on information acquired by several environmental sensors located at different positions on the vehicle. In other words, information from different environmental sensors located at different positions on the vehicle can be used in parallel to change the occupancy values ​​of cells and, after the occupancy values ​​have changed, to obtain the resulting occupancy information. This allows an environmental image to be generated or an existing environmental image to be updated using multiple sensor information items in parallel.

[0029] Steps a) to d) can, in particular, be performed multiple times in succession. The environmental information determined using the method according to the invention can be used, for example, within a driver assistance system, in particular for detecting and / or assisting in approaching parking spaces in the vicinity of the vehicle. It is particularly possible for the vehicle to move between the transmission of multiple sensor signals.

[0030] It is possible to compensate for the ego movement of the vehicle that occurs between the transmission and reception of the sensor signal. This makes it possible to compensate for the change in the vehicle's position relative to a stationary object in the vehicle's surroundings that occurs between the transmission and reception of the sensor signal. The detection of environmental objects at different vehicle positions can be considered environmental detection using a synthetic aperture.

[0031] In one embodiment of the invention, it is possible to also evaluate cross echoes between different ultrasonic sensors to determine the occupancy information. In other words, a reflected signal component received by an ultrasonic sensor other than the transmitting sensor also contributes to the determination of the occupancy information and thus to an improved determination of the environmental information.

[0032] According to one embodiment, the grid is a grid that moves with the vehicle. The grid can, for example, have a length and / or width corresponding to 3 to 6 times the length of the vehicle and, for example, have an area of ​​20 m by 20 m. Thus, the grid encompasses the immediate surrounding area of ​​the vehicle, which can be detected by environmental sensors provided on the vehicle, such as ultrasonic sensors. This advantageously allows for savings in memory and computing resources, particularly in a control device designed to implement the method.

[0033] According to the invention, a complex count value can be used as the count value that depends on the angle between the respective cell and the current position of the environmental sensor. Thus, a count value comprising a real part and an imaginary part can be used, which can represent the angle between the respective cell and the current position of the environmental sensor. The occupancy value can also be a complex number consisting of a real part and an imaginary part.

[0034] According to the invention, the real and imaginary parts of the count value are added separately for each cell, with the occupancy value of the cell corresponding to the value of the complex number formed from the added real and imaginary parts. Since the real and imaginary parts can become negative depending on the angle between a cell and the current sensor position for point-shaped objects, the overall occupancy value for point-shaped objects is lower when added together. This allows for an approximation to the occupancy values ​​that are usually determined for linear objects.

[0035] In a preferred embodiment of the invention, it can be provided that a count value dependent on the speed of the vehicle, in particular a count value multiplied by a speed-dependent factor, is used as a count value dependent on the detection intensity of the respective cell. The magnitude of the speed-dependent count value can preferably be proportional to the speed of the vehicle. Consequently, the occupancy value of a cell is changed by a lower count value for a slowly moving vehicle than for a fast-moving vehicle. The speed-dependent count value can compensate for the effect that individual cells are detected more frequently when the vehicle is traveling slowly, since they are more often within the detection range of the sensor.When a vehicle is moving faster, individual cells can move out of the detection range of the environment sensor more quickly, so they are generally detected less frequently. This effect can be compensated for by taking the vehicle speed into account when calculating the count value, which is used to change the occupancy value. This advantageously ensures that the objects in the environment map appear with a similar level of definition. By summing the count values, excessively high occupancy values ​​for objects detected when the vehicle was moving slowly can be avoided. This advantageously improves the raster map and thus also the information about the environment.

[0036] The use of a speed-dependent factor, which is multiplied by a fixed increment or an otherwise variable count value, offers a simple implementation of the speed dependency of the count value. In particular, this offers the possibility of easily multiplying a fixed increment or an already variable count value by several factors in order to map multiple dependencies of the count value and thus form the variable count value. In particular, a combination of a speed-dependent factor with a complex-valued count value and / or with factors dependent on the detection intensity of the cell, as described below, can be provided.

[0037] According to the invention, it can be provided that a count value dependent on a detection frequency of the cell, in particular a count value multiplied by a factor dependent on the detection frequency, is used as a count value dependent on the detection intensity of the respective cell. The count value dependent on the detection intensity can preferably be inversely proportional to a detection frequency assigned to the respective cell. In other words, the count value can become increasingly smaller the more frequently the cell has already been detected. To determine the detection frequency, each cell can be assigned detection frequency information which indicates how often a reflected signal component has already been assigned to the respective cell. The detection frequency of a cell can, for example, initially be zero and can be increased by an increment of one with each assignment of a signal component.

[0038] Taking detection frequency into account means that cells that are detected frequently experience a smaller change in their occupancy value than cells that are detected less frequently. This allows the occupancy values ​​of closer and more distant objects to be advantageously aligned. In addition, differences in detection frequency due to changes in vehicle speed can also be compensated for.

[0039] According to the invention, it can be provided that a count value dependent on a detection duration, in particular a count value multiplied by a factor dependent on the detection duration of the respective cell, is used as a count value dependent on the detection intensity of the respective cell. Preferably, the count value dependent on the detection intensity can be inversely proportional to a detection frequency assigned to the respective cell. For this purpose, for example, it can be stored for each cell how long, i.e. for which period of time, the respective cell was in the detection range of the environmental sensor, with the count value decreasing as the detection duration increases. The effect of taking the detection duration into account essentially corresponds to that of taking the detection frequency into account.

[0040] In a preferred embodiment of the invention, a count value dependent on a reflection density, in particular a count value multiplied by a factor dependent on the reflection density of the respective cell, can be used as a count value dependent on the detection intensity of the respective cell. The reflection density can, in particular, describe the ratio between the number of reflected signal components assigned to the respective cell and the number of sensor signals transmitted in the direction of the respective cell. A factor dependent on the reflection density can be proportional to the reflection density.

[0041] The number of sensor signals transmitted toward the respective cell can be stored as separate information for each cell, for example, in the form of an additional raster map containing this number as the occupancy value of the individual cells. The number of sensor signals transmitted toward one of the cells can be increased for all cells located within a sensor field of view and within the sensor visibility of the environmental sensor when the sensor signal is transmitted.

[0042] In addition to or as an alternative to an angle-dependent count value and / or a count value dependent on the detection intensity, further dependencies of the count value can also be provided. In particular, further factors, which depend, for example, on the amplitude of the reflected signal component, the azimuth angle of the reflected signal component, and / or similar factors, can also be used and applied to the variable count value.

[0043] According to the invention, the occupancy value of one or more cells can be compared with a threshold value, and if the threshold value is exceeded, the occupancy status of the cell is set to occupied. A cell with an occupancy status describing the occupancy of the cell describes a spatial section in the vehicle's surroundings that is blocked by an object or part of an object.

[0044] Furthermore, according to the invention, each of the cells can be assigned a detection frequency information item describing the number of previously assigned signal components, with those cells for which the detection frequency information exceeds a further limit value also being marked as occupied. This can prevent a situation where, when completely circling a point-like object using an angle-dependent count value, in particular a complex count value, such a cell is incorrectly considered unoccupied due to the at least partial cancellation of the summands forming the occupancy value caused by sweeping the entire angular range from 0 to 360°.In other words, cells that have no or only a low occupancy value can also be marked as occupied if they have already been assigned a number of reflected signal components specified by the further limit value.

[0045] Whether it is necessary to assign a detection frequency to the cells can depend largely on the scenario in which the surrounding information is to be determined. When using the surrounding information for parking space detection, for example, in a driver assistance system, completely avoiding objects is very unlikely, so using the detection frequency information is not absolutely necessary.

[0046] The invention further relates to an environment detection device comprising a sensor device with at least one environment sensor and a control device, wherein the control device is configured to carry out a method according to the invention.

[0047] Furthermore, the invention relates to a vehicle with an environment detection device according to the invention.

[0048] The invention further relates to a computer program product containing instructions that cause a computing device of a system comprising the computing device and a sensor device with at least one environmental sensor to carry out a method according to the invention. The computing device can, in particular, be a control device of an environmental detection device of a vehicle. All advantages and configurations described above with reference to the method according to the invention apply accordingly to the environmental detection device according to the invention, to the vehicle according to the invention, and to the computer program product according to the invention, and vice versa.

[0049] The advantages and embodiments described in relation to the environment detection device according to the invention, the vehicle according to the invention and the computer program product according to the invention are also transferable to one another.

[0050] Further advantages and embodiments of the invention will become apparent from the exemplary embodiments described below and from the drawings. These are schematic representations and show:

[0051] Fig. 1 shows an embodiment of a vehicle according to the invention comprising an embodiment of an environment detection device according to the invention,

[0052] Fig. 2 is a flowchart of an embodiment of a method according to the invention,

[0053] Fig. 3 shows an example of environmental information created by means of an embodiment of the method according to the invention,

[0054] Fig. 4 shows an example of environmental information generated by a method according to the prior art,

[0055] Fig. 5 shows the environmental information shown in Fig. 3 after applying a threshold value to determine cells considered occupied,

[0056] Fig. 6 shows the environmental information shown in Fig. 4 after applying the threshold value to determine cells considered to be occupied, Fig. 7 shows the environmental information shown in Fig. 4 after applying a further threshold value to determine cells considered to be occupied,

[0057] Fig. 8 a detection of a point-like object by a passing vehicle and

[0058] Fig. 9 a detection of a linear object by a passing vehicle.

[0059] Fig. 1 shows an embodiment of a vehicle 1. The vehicle 1 can be, for example, a passenger car, a truck, or another motor vehicle. It is possible for the vehicle 1 to be a rail-bound vehicle and / or for the vehicle 1 to be a non-motorized vehicle, for example, a trailer or the like, or to include such a non-motorized vehicle as a combination.

[0060] The vehicle 1 comprises an exemplary embodiment of an environment detection device 2, which comprises a control device 3 and a plurality of environment sensors 4. The environment sensors 4 are designed, for example, as ultrasonic sensors. Alternatively, the environment sensors 4 can be designed as radar sensors, lidar sensors, or similar. In the present exemplary embodiment, the vehicle 1 comprises twelve environment sensors 4, with six environment sensors arranged at the front of the vehicle 1 and six environment sensors arranged at the rear of the vehicle 1. The number and / or arrangement of the environment sensors 4 is purely exemplary.

[0061] Using the environmental sensors 4, sensor signals, in this case ultrasonic signals, can be transmitted into the surroundings of the vehicle 1. Signal components reflected by objects in the surroundings of the vehicle 1 can be received again by one or more of the environmental sensors 4. The reflected signal components can also be referred to as echoes.

[0062] The control device 3 of the environment detection device 2 is designed to carry out a method for generating environmental information describing the surroundings of the vehicle 1. The environmental information comprises a raster map comprising a plurality of cells. The raster map can be created or provided, for example, as a data structure in a memory device of the control device 3.

[0063] The cells of the raster map are each assigned to a sub-area of ​​the surrounding area of ​​vehicle 1, and each cell is assigned an occupancy value. Each cell describes a section of a spatial area around vehicle 1. The cells can be rectangular, for example, although other geometric shapes are also possible. The raster map can, for example, have dimensions of 20 m by 20 m, whereby the cells can have lengths and widths in the centimeter range, depending on the desired resolution of the raster map.

[0064] Fig. 2 shows an embodiment of the method for determining environmental information. Step S1 involves the transmission of a sensor signal by at least one of the environmental sensors 4. The ultrasonic signal transmitted into the environment of the vehicle 1 can be reflected by objects, with a reflected signal component subsequently being reflected back toward the vehicle 1 as an echo and being received by at least one of the environmental sensors 4.

[0065] Subsequently, in a step S2, the ultrasonic signal or the reflected signal component reflected from an object is received again by one of the environmental sensors 4 of the environmental detection device 2 or of the vehicle 1. The reflected signal component can be received by the same environmental sensor 4 that emitted the ultrasonic signal, and / or the reflected signal component can also be received by another of the environmental sensors 4 of the vehicle 1.

[0066] In step S3, the reflected and received signal portion is assigned to one or more of the cells of the raster map based on a current position of the receiving environmental sensor 4. The position of the environmental sensor 4 may differ upon reception from a position upon transmission, in particular if the vehicle 1 is moving, i.e. has moved in the time between the transmission of the transmission signal and the reception of the reflected signal portion.

[0067] The allocation of the reflected and received signal portion to one or more of the cells is carried out, in particular, depending on a distance value described by the reflected signal portion, which describes the distance between the environmental sensor 4 and the object at which the previously transmitted sensor signal was reflected. It is possible that the movement of the vehicle between the transmission of the sensor signal and the reception of the corresponding echo is taken into account in this spatial allocation.

[0068] A received, reflected signal component, to which a specific distance value is assigned, can be assigned to all cells located within a radius around the ultrasonic sensor corresponding to the distance value. The assignment can be limited to cells that are generally within the detection range of the environmental sensor 4, and thus, for example, along a circular segment.

[0069] In step S4, the occupancy value of those cells to which the reflected signal component was previously assigned is subsequently changed. The occupancy value is changed depending on a variable count value, whereby the count value depends on an angle between the respective cell and the current position of the ambient sensor and / or on a detection intensity assigned to the respective cell. The occupancy value is changed, in particular, by adding the count value to the previously existing occupancy value, whereby an initial occupancy value of zero can be assumed for each of the cells.

[0070] A complex count value can be used as a count value that depends on the angle between the respective cell and the current position of the environmental sensor 4. The occupancy value m(x,y)t of a cell can be calculated according to the following formula: (cos 0x ,y,x,y + j sin Qx,y,x,y ) (1 ) where mx,y _ the previous occupancy value of the cell, j the complex unit and cos 0 x ,y,x,y describes the angle between the respective line and the current sensor position.

[0071] As angle 0 x ,y,x,y, for example, the angle between a fixed x-axis as a reference and a vector based on the current sensor position (x,y) and the position of the respective cell can be used. The angle d x ,y,x,y as be calculated.

[0072] The use of such an angle-dependent count value to calculate the occupancy value of the cells of the raster map has the advantage that for a linear object the same angle 6 x,y,x,y is present, so that the count values ​​increase in absolute value when added up at about the same rate as with point-like objects which are detected at a changing angle. To calculate the occupancy value, the real part and the imaginary part of the angle-dependent count value can each be added separately. With point-like objects which are continuously detected at different angles while vehicle 1 is moving, the effect occurs that, depending on the angle, the real part and / or the imaginary part can become negative, so that when the corresponding real or imaginary part is added up, a reduction in the absolute value of the occupancy information can also occur. This significantly leads to the occupancy values ​​or the absolute values ​​of the complex-valued sums of the count values ​​increasing less than when a constant and non-changing increment is used as the count value.

[0073] It is possible that in a subsequent step S5 (not shown in Fig. 2) a comparison of the respective occupancy value of the cells of the raster map is carried out with a limit value, whereby all those cells are marked as occupied by an environmental object for which the occupancy value, in particular the amount of the occupancy value, exceeds a predetermined limit value.

[0074] To avoid receiving occupancy information of 0 when completely circling or encircling a point-shaped object, where the object is detected in the entire angular range between 0° and 360°, detection frequency information can also be stored for the individual cells. This information indicates how often a reflected signal component has already been assigned to the corresponding row. Cells where the occupancy value is below the threshold value, but which have a detection frequency above another threshold value, can be considered cells containing a point-shaped object and marked as occupied accordingly.

[0075] In addition to or as an alternative to using the angle-dependent count value, a count value dependent on the detection intensity of the respective cell can also be used. For example, a count value can be used that depends on the current speed of vehicle 1. A count value dependent on a detection intensity can, in particular, be a factor T dependent on the detection intensity, which is multiplied by a constant value to form the count value. In particular, it is also possible for a factor T dependent on the detection intensity to be multiplied by a complex count value, which, as described above, depends on the angle between the respective cell and the current position of the environmental sensor. In this case, an occupancy value is determined according to the following formula: (cos 0 x ,y,x,y + j sin 9 x.yx,y ) * T (3) where the factor T depends, for example, on the speed of the vehicle. It is possible that the factor T can be, for example, the speed of vehicle 1 according to is used. Alternatively, other types of speed dependence of the factor T are also possible.

[0076] The advantage of using the factor dependent on the speed of vehicle 1 is that, when vehicle 1 is moving slowly, only small count values ​​are added to form the occupancy value, which compensates for the fact that, when vehicle 1 is moving slowly, objects in the vicinity of vehicle 1 are detected more often than if vehicle 1 were moving faster.

[0077] The effect of the variable count value when the occupancy information of the cells of a raster map determined as environmental information changes will be illustrated by the following example.

[0078] Fig. 3 shows environmental information which was determined using formula (3). For comparison, Fig. 4 shows environmental information based on the same sensor data, but which was determined with a constant, i.e. non-variable, count value c according to the formula m(x,y) = m(x,y) + c (5). Fig. 3 and Fig. 4 each show the occupancy values ​​assigned to the cells as gray levels, with the occupancy values ​​resulting from the sum of the respectively added count values. The count values ​​were formed from reflected sensor signals which were received during an at least temporary movement of the vehicle 1 along a parking space 5 between two other vehicles 6, 7. Between the other vehicles 6, 7, the parking space 5 is still bordered on one side by a curb 8.

[0079] As can be seen in the areas 9 and 10, each marked with a rectangle and corresponding to a corner of one of the other vehicles 6, 7, the method according to the invention leads to a reduced detection of cells that are falsely positively marked as occupied. The contours of the other vehicles 6, 7 bordering the parking space 5 and the curb 8 are more clearly visible. By using the speed-dependent count value, which in this case is formed by using the speed itself as a speed-dependent factor, it can be achieved that fewer false positive detections occur in area 11.

[0080] The semicircularly arranged cells in area 11 in Fig. 4 are based on reflected sensor signals received when the vehicle 1 is stationary, i.e., at a speed of zero. When calculating the occupancy values ​​of the cells according to formula (3), as can be seen in Fig. 3, the occurrence of high occupancy values ​​in this area is prevented by using the speed-dependent factor.

[0081] As shown in Fig. 5 to Fig. 7, the calculation of the occupancy values ​​according to formula (3) also affects the determination of the occupancy state of the individual cells. As can be seen from Fig. 5 to 7, using a single limit value for occupancy values ​​determined according to formula (3) results in a much clearer picture of the vehicle environment (Fig. 5) than when calculating the occupancy values ​​according to formula (5) (Figs. 6 and 7), whereby in Figs. 6 and 7 the occupancy states were each determined using a different limit value. The limit values ​​each indicate the amount of occupancy information above which a cell is marked as occupied (black in the figures). Cells with an amount of occupancy information below the limit value are considered unoccupied (white in the figures).

[0082] The threshold value for determining the occupancy states, as shown in Fig. 5, was chosen to minimize the number of false-positive detections without removing any actual detections, so that the number of cells with false-negative occupancy states is also as low as possible. Using the same threshold value for the occupancy states determined using formula (5), a high number of false-positive occupancy states results. This is clearly evident in Fig. 6, where the same threshold value was used to determine occupied cells as in the evaluation underlying Fig. 5.

[0083] If the threshold is reduced in order to reduce the number of cells falsely identified as occupied for occupancy states determined according to formula (5), the number of false-negative detections increases, as can be seen in Fig. 7. The contours of the other vehicles 6, 7 and the curb 8 are no longer as clearly visible, since they partly run along cells incorrectly marked as unoccupied.

[0084] In addition or alternatively to a speed-dependent count value, a count value can also be used which depends on the detection intensity of the respective cell. The detection intensity can in particular describe the detection frequency and / or the detection duration, wherein the detection frequency describes the number of reflected signal components previously assigned to the respective cell and the detection duration describes the time in which the cell is located in the detection range of the environmental view 4. A count value dependent on the detection intensity can be formed by a factor T which depends on the detection intensity. The factor can in particular be inversely proportional to the detection intensity. For example, a factor

[0085] T (x,y) = l / n (6) can be used, where n describes the number of reflected signal components previously assigned to the respective cell.

[0086] Analogously, a factor T determined inversely proportional to the detection duration can also be used. With such detection intensity-dependent count values, results comparable to the speed-dependent count value can be achieved. In particular, the factors can also be combined with a complex count value to achieve uniform occupancy values ​​for point-like and linear objects.

[0087] In a further embodiment, a count value dependent on a reflection density can be used as a count value dependent on the detection intensity of the respective cell. For this purpose, a factor T can be used that is proportional to a reflection density of the respective cell, wherein the reflection density describes the ratio between the number of reflected signal components assigned to the respective cell and the number of sensor signals transmitted in the direction of the respective cell.

[0088] The factor T can be defined as where #TR describes the number of reflected signal components assigned to the respective cell and #Tx describes the number of sensor signals transmitted in the direction of the respective cell. The number of sensor signals transmitted in the direction of the respective cell can be stored as separate information for each cell of the raster map, for example in the form of another raster map containing this number as the occupancy value of the individual cells, or in each case as a number information item describing this number and assigned to each of the cells.

[0089] The number of sensor signals transmitted in the direction of one of the cells can be increased when the sensor signal is transmitted in step a) of the method for all those cells of the raster map which lie within a sensor field of view of the environmental sensor 4, i.e. within an azimuth angle range, and within a predetermined sensor field of view of the transmitting environmental sensor 4.

[0090] Fig. 8 shows a situation in which a point-like object 12 is detected by the environmental sensor 4 from different sensor positions pi. This situation can occur, for example, when the vehicle 1 is moving along the x-axis.

[0091] For each of the sensor positions pi, a circle 13 is shown with a radius corresponding to the determined distance between the environmental sensor 4 and the object 12. In particular, those cells in the raster map that are intersected by the circle shown experience a change in their occupancy value. All circles 13 clearly intersect at a point P, so that at least one cell encompassing this point has a high number of assigned reflected signal components or changes in its occupancy value.

[0092] Fig. 9 shows a situation in which a linear object 14 is detected by the environmental sensor 4 from different sensor positions pi, although this situation can also occur when the vehicle 1 is moving along the x-axis. For each of the sensor positions pi, a circle 13 is shown with a radius corresponding to the determined distance between the environmental sensor 4 and the object 14, whereby in this case too, in particular those cells in the raster map which are intersected by the shown circle experience a change in their occupancy value. It is evident that the circles intersect at a plurality of points P, each of which is located slightly in front of the linear object 14. This means that the cells comprising these points experience a change in their occupancy value more often than the cells through which the linear object 14 passes.

Claims

Patent claims 1. A method for generating environmental information describing the surroundings of a vehicle (1), wherein the vehicle (1) has a sensor device with at least one environmental sensor (4) and the environmental information comprises at least one raster map, wherein the raster map has a plurality of cells, each of which is assigned to a sub-area of ​​the surrounding area of ​​the vehicle and to which an occupancy value is each assigned, comprising the steps of: a) transmitting a sensor signal by the environmental sensor (4); b) receiving a reflected signal component of the sensor signal by the environmental sensor (4); c) assigning the reflected signal component to one or more cells of the raster map based on a current position of the environmental sensor (4);d) changing the respective occupancy values ​​of the cells to which the reflected signal component is assigned by a variable count value, wherein the count value depends on an angle between the respective cell and the current position of the environmental sensor (4) and / or on a detection intensity assigned to the respective cell; 2. Method according to claim 1, characterized in that a complex count value is used as the count value depending on the angle between the respective cell and the current position of the environmental sensor (4).

3. Method according to claim 2, characterized in that for each cell the real part and the imaginary part of the count value are added separately, the occupancy value of the cell corresponding to the amount of the added count values.

4. Method according to one of the preceding claims, characterized in that a count value dependent on the speed of the vehicle (1), in particular a count value multiplied by a speed-dependent factor, is used as a count value dependent on the detection intensity of the respective cell.

5. Method according to claim 4, characterized in that the magnitude of the speed-dependent count value is proportional to the speed.

6. Method according to one of the preceding claims, characterized in that a count value dependent on a detection frequency of the cell, in particular a count value multiplied by a factor dependent on the detection frequency, is used as a count value dependent on the detection intensity of the respective cell.

7. Method according to one of the preceding claims, characterized in that a count value dependent on a detection duration, in particular a count value multiplied by a factor dependent on the detection duration of the respective cell, is used as a count value dependent on the detection intensity of the respective cell.

8. Method according to one of the preceding claims, characterized in that a count value dependent on a reflection density, in particular a count value multiplied by a factor dependent on the reflection density of the respective cell, is used as a count value dependent on the detection intensity of the respective cell.

9. Method according to one of the preceding claims, characterized in that the count value dependent on the detection intensity is inversely proportional to a detection frequency and / or a detection duration of the cell, and / or that the count value dependent on the detection intensity is proportional to a reflection density of the cell.

10. Method according to one of the preceding claims, characterized in that the amount of the occupancy value of one or more cells is compared with a limit value, wherein if the limit value is exceeded, an occupancy state of the cell is set to occupied.

11. Method according to claim 10, characterized in that each of the cells is assigned detection frequency information describing the number of previously assigned signal components, wherein in addition those cells are also marked as occupied for which the detection frequency information exceeds a further limit value.

12. Environmental detection device comprising a sensor device with at least one environmental sensor (4) and a control device (3), wherein the control device (3) is configured to carry out a method according to one of the preceding claims.

13. Vehicle comprising an environment detection device (2) according to claim 12.

14. A computer program product containing instructions to cause a computing device of a system comprising the computing device and a sensor device with at least one environmental sensor to carry out a method according to one of claims 1 to 11.

Images