A method for characterizing objects in the surrounding environment of an automobile.

Abstract

The present invention relates to a method (100) for characterizing an object in the surrounding environment of an automobile by an automobile assistance system, wherein the automobile is moved relative to the object, a measurement signal is emitted by a distance sensor of the assistance system, an echo of the measurement signal reflected by the object is received, and a control device of the assistance system is used to determine the height classification of the object based on an echo density parameter determined based on the number of echoes received. According to the present invention, a curve of the echo density parameter is determined, and the height classification of the object is determined based on the curve of the echo density parameter. The present invention also relates to an assistance system including an ultrasonic sensor and a control device designed to perform such a method (100).

Description

Technical Field

[0001] The present invention relates to a method for characterizing an object in the surrounding environment of a motor vehicle by means of an assistance system of the motor vehicle, wherein the motor vehicle is moved relative to the object, a distance sensor of the assistance system is used to emit a measurement signal, and an echo of the measurement signal reflected by the object is received. In this context, a control device of the assistance system is used to determine a classification of the height of the object based on an echo density parameter determined based on the number of received echoes. The present invention also relates to an assistance system comprising a distance sensor and a control device designed to carry out such a method.

Background Art

[0002] Distance sensors in motor vehicles are usually used to detect the surrounding environment, and the distance, i.e., the distance to an object, is determined based on the transit time between the emitted measurement signal and the received echo of the measurement signal reflected by the object. Such distance sensors are generally designed as radar sensors or ultrasonic sensors, and in particular ultrasonic sensors are widely used and are currently installed in almost all passenger cars.

[0003] An ultrasonic sensor usually comprises a transmitter that emits an ultrasonic signal as a measurement signal that propagates in air at a speed of sound of approximately 340 meters per second. For this purpose, the membrane of the ultrasonic sensor is usually excited by a corresponding transducer to execute mechanical vibrations. The ultrasonic signal is reflected as an echo by an object in the surrounding environment and detected by the receiving device of the ultrasonic sensor. Based on the transit time difference between the transmission time and the reception time, taking into account the propagation speed of the ultrasonic signal, the distance or the distance to the object may be determined. In this case, the amplitude of the reflected ultrasonic signal or the amplitude of the echo may also be determined.

[0004] Automotive ultrasonic sensors are typically used to detect the surrounding environment within a range of up to approximately 7 meters. Ultrasonic sensors are particularly important in parking applications, especially in semi-automatic or automatic driving operations, particularly in relation to measuring parking distance, searching for parking spaces, or parking itself. In this case, the vehicle is usually moving relative to an object, and one measurement cycle is performed for a predetermined time during that movement in each case. In this context, an ultrasonic signal is emitted by the ultrasonic sensor during each measurement cycle. In the prior art, methods and corresponding assistance systems are known that use ultrasonic sensors to provide the driver with various information about the surrounding environment of a vehicle, assisting the driver in operating the vehicle, and in particular in searching for and parking a parking space. For example, there are assistance systems with a parking space search function that indicate to the driver whether a parking space exists in the surrounding environment very close to the vehicle, or whether an existing parking space is large enough to accommodate the vehicle. To reliably identify and measure a parking space, such assistance systems require information about objects located within the surrounding environment of the vehicle, which can be formed, for example, by parked vehicles, curbs, walls, and stonework.

[0005] In addition to the distance from the object to the vehicle, the object's height is usually also important. Height is a crucial factor in determining whether an object or obstacle can pass through. It is particularly desirable to determine the height of acquired objects when the vehicle is operated at least semi-autonomously based on distance sensor measurements.

[0006] Determining height using one-dimensional (1D) distance sensors, i.e., distance sensors commonly used in the automotive sector for distance determination, is fundamentally very difficult due to physical limitations. It is impossible to directly measure the height of an object using such distance sensors. Therefore, in determining height, methods are employed that involve additionally using, for example, a camera to estimate height based on 2D images, or a multi-sensor-based method to estimate height based on triangulation. However, camera-based or multi-sensor-based methods do not take advantage of the benefits of 1D distance sensors in terms of cost and robustness.

[0007] The type of method and support system described at the beginning is known, for example, from DE102007039348A1. In this regard, an ultrasonic signal is emitted by an ultrasonic sensor in a vehicle, and echoes formed as ultrasonic pulses of the ultrasonic signal reflected by an object are received. For the received ultrasonic pulses, a reference point is determined as a point in time between a first point in time when the ultrasonic pulse exceeds a first threshold and a subsequent second point in time when the ultrasonic pulse falls below a second threshold, and the distance to the object is determined based on this determined reference point. Furthermore, the classification of the object's height is determined based on the pulse width of the reflected received ultrasonic pulse, the ratio of the number of received ultrasonic pulses to the number of emitted ultrasonic pulses, and / or the distribution of distances determined by the travel time measurements from multiple reflected ultrasonic pulses.

[0008] DE102004047479A1 discloses a further method for characterizing objects in the surrounding environment of an automobile. In this method, ultrasonic signals are emitted by an ultrasonic sensor to classify the height of an object as the automobile passes it relative to the side of the automobile. The classification of the object's height is determined based on a comparison of the amplitude of the received echo with a predetermined possible threshold. [Prior art documents] [Patent Documents]

[0009] [Patent Document 1] DE102007039348A1 [Patent Document 2] DE102004047479A1 [Overview of the project] [Problems that the invention aims to solve]

[0010] This invention aims to define an alternative method for characterizing objects in the surrounding environment of an automobile, and a corresponding support system. This method makes it possible to classify the height of objects as inexpensively, accurately, and reliably as possible. [Means for solving the problem]

[0011] This objective is achieved by the teachings of claim 1 and independent claim 12 as a whole. Suitable embodiments and improvements of the present invention are described in the dependent claims and the following description.

[0012] The present invention relates to a method for characterizing objects in the surrounding environment of an automobile using an automobile assistance system. In this method, the automobile is moved relative to the object, a distance sensor of the assistance system emits a measurement signal, and an echo of the measurement signal reflected by the object is received. In connection therewith, a control device of the assistance system is used to determine the height classification of the object based on an echo density parameter determined based on the number of echoes received.

[0013] According to the present invention, an echo density parameter curve is determined, and the classification of the object's height is determined based on the echo density parameter curve.

[0014] In this case, the present invention is based on the consideration that it is necessary to regularly use absolute or fixed thresholds or limits, particularly predefined thresholds or limits for specific boundary conditions or situations, for the purpose of determining classification, which may periodically lead to inaccurate or erroneous classification results. This is because, for example, the current temperature and humidity in the environment surrounding the vehicle each have a significant effect on the attenuation of sound in the air, and the current speed of the vehicle has a significant effect on the number of reflected echoes, the number of receivable echoes, or the number of echoes received. In contrast, using thresholds or limits that are adaptable during the vehicle's operation, such as the current temperature, current air humidity, and / or current speed of the vehicle, results in a very high evaluation and computational load. Furthermore, the number of reflected echoes, or echoes that can be received, or are generally received, also depends on the material of the object or the reflectivity of the material. This is usually unknown and cannot be determined, and therefore cannot be used to define or apply thresholds or limits. Therefore, the present invention is based on the consideration that a particularly cost-effective and reliable classification of object heights can be achieved, in particular, by minimizing the use of thresholds or limits for comparison with echo density parameters or variables derived therefrom, especially at low cost and with reliable classification. Accordingly, according to the present invention, a curve of echo density parameters is determined, and the classification of object heights is determined based on the curve of echo density parameters.

[0015] The design according to the present invention has the advantage of providing a cost-effective, accurate, and reliable method for classifying the height of objects.

[0016] The objects being characterized may be stationary objects that extend from the ground, for example, a road surface or other terrain, and that extend substantially perpendicular to that ground, such as curbs, walls, or vehicles. However, they may also be objects that do not extend from the ground, such as fence rails, or objects that do not extend perpendicular to the ground, such as ramps.

[0017] The car, advantageously, moves towards or away from an object. This causes the car to approach or move away from the object. During the corresponding movement of the car, the echo density parameter is determined intermittently or continuously.

[0018] The echo density parameter curve is, advantageously, a time-dependent curve and / or distance-dependent curve, i.e., a curve that depends on the distance (or distance) from the object to the vehicle.

[0019] Distance sensors can be placed in different locations, such as in front of, behind, or to the side of a vehicle, and are advantageously implemented as 1D distance sensors. It is also possible to use only a single distance sensor. However, it is also possible to use multiple distance sensors.

[0020] In particular, two classes, "high" and "low," are used to classify the height of an object. Here, an object is classified as "high" if it is an object that a vehicle cannot pass over, especially an object related to collision, and an object is classified as "low" if it is an object that a vehicle can pass over. Alternatively, an object is classified as "high" if it is located at least the installation height of the distance sensor, that is, if the object has a height that corresponds to at least the installation height of the distance sensor. An object is classified as "low" if it is located below the installation height of the distance sensor, that is, if the object has a height that is lower than the installation height of the distance sensor.

[0021] In one advantageous embodiment, no comparison is made between the echo density parameter curve and, in particular, a predetermined threshold, in determining the classification of the object's height. The echo density parameter curve is not compared to, in particular, a predetermined threshold or limit, and therefore no comparison to a threshold or limit is made herein. Thus, the corresponding determination is made regardless of whether it is above or below the threshold or limit. Advantageously, in determining the classification of the object's height, the determined echo density parameter is also not compared to, in particular, a predetermined threshold or limit, and therefore the determination is made regardless of whether it is above or below the threshold or limit.

[0022] In a more advantageous embodiment, the height classification of an object is determined based on a change in the curve of the echo density parameter, preferably an increase and / or decrease. Thus, the height classification of an object is determined based on a change in the curve, in particular on distance, or the distance from the object to the vehicle or distance sensor. In this case, the height classification of an object is preferably determined based on an increase and / or decrease in the curve, in particular on whether there is an increase or decrease in the curve of the echo density parameter. In accordance with this determination, the classification of the object is determined, and in particular the object is classified as "high" or "low".

[0023] In more advantageous embodiments of this specification, an object is located near a vehicle, preferably at a distance of up to 3 meters, more preferably up to 2 meters, from the vehicle's distance sensor, and is classified as low if a downward curve of the echo density parameter is determined as the vehicle approaches the object, and as high if an upward curve of the echo density parameter is determined. In particular, in this specification, an object is classified as "low" if it is located below the installation height of the distance sensor, i.e., if the object has a height particularly lower than the installation height of the distance sensor. Such objects are, for example, curbs. In particular, an object is classified as "high" if it is located at least at the installation height of the distance sensor, i.e., if the object has a height particularly corresponding to at least the installation height of the distance sensor. Such objects are, for example, walls, fences, or vehicles.

[0024] This is based, in particular, on the fact that objects at a height corresponding at least to the installation height of the distance sensor in the vehicle have a higher reflectivity with respect to the measurement signal than objects at a height lower than the installation height of the distance sensor. Thus, the measurement signal from an object having a height corresponding at least to the installation height of the distance sensor in the vehicle is reflected by a plurality of points on the surface area of an object positioned orthogonally and non-orthogonally to the distance sensor. In contrast, the measurement signal from an object at a height lower than the installation height of the distance sensor is reflected only by points on the surface area of an object not positioned orthogonally to the distance sensor. This results in a reduction in the number of reflected measurement signals or echoes. Furthermore, the radiation pattern of the distance sensor, particularly an ultrasonic sensor, is basically a function of the elevation angle. That is, the power of the measurement signal emitted by the distance sensor to an object within the detection range depends on the elevation angle of the object with respect to the distance sensor. In the case of an object having a height lower than the installation height of the distance sensor in the vehicle, as a result, its elevation angle, and thus the power, or rather the amplitude of the reflected measurement signal, changes. In particular, when the distance between the object and the vehicle becomes shorter than a predetermined distance, more precisely, when the distance between the object and the distance sensor falls below the predetermined distance, it occurs according to the distance between the vehicle or the distance sensor and the object. In this case, as the distance becomes shorter, i.e., as the vehicle approaches the object further, the elevation angle decreases, which may lead to a reduction in the amplitude of the reflected measurement signal to such an extent that the measurement signal reflected by the object is not detected.

[0025] In a further advantageous embodiment, the classification of the height of the object is determined based on the slope of the curve of the echo density parameter. Thus, in this case, for the classification of the height of the object, the slope or slopes of the curve of the echo density parameter are evaluated, and based thereon, the height of the object is classified. In this case, the slope of the curve of the echo density parameter is, in particular, the (first-order) time derivative of the curve of the echo density parameter. Based on the determined slope, the height of the object can be determined more precisely, or more accurately, or more discriminatively. In this case, as the classification of the height of the object, in addition to or instead of the two classes of "high" and "low", more than two classes and / or more subdivided classes, for example, "high, impassable", "low, impassable", "low, passable", and "low, can be opened without damaging the vehicle door" can be used.

[0026] Advantageously, in this case, the object is located near the motor vehicle, preferably at a distance of at most 3 meters, more preferably at most 2 meters, from the distance sensor of the motor vehicle. When the slope of the curve of the echo density parameter is determined to be negative when the motor vehicle approaches the object, the object is classified as "low", and when the slope of the curve of the echo density parameter is determined to be positive, the object is classified as "high". Preferably, a more detailed or more discriminative classification of the height of the object is also performed in this case based on the determined (numerical) value of each determined slope. In particular, more than two classes and a classification simply distinguished as "high" and "low" are used.

[0027] The echo density parameter can, in particular, correspond to the number of received echoes.

[0028] However, in a more advantageous embodiment, the echo density parameter is determined as the ratio of the number of echoes received to a predetermined time measurement window, where the number of echoes received is preferably normalized with respect to the duration of the measurement window. Thus, the number of echoes is preferably normalized to a predetermined duration, which in particular corresponds to the time during which the corresponding object is detected by the distance sensor, i.e., the time during which the measurement signal is emitted to the object by the distance sensor. For example, if the time measurement window for a first object is larger than the time measurement window for a second object, where the area of ​​the first object detected or detectable by the distance sensor, particularly the surface area, is the same size as the area of ​​the second object detected or detectable by the distance sensor, particularly the surface area, and the same number of echoes are received from the first object as from the second object, the echo density parameter for the first object is lower than that for the second object.

[0029] Alternatively or additionally, in further advantageous embodiments, the echo density parameter is determined as the ratio of the number of received echoes to a given area in the surrounding environment, and the number of received echoes is preferably normalized with respect to the given area in the surrounding environment. Thus, the number of echoes is preferably normalized with respect to a given surrounding area or a given area unit. The given area is preferably the area, particularly the surface area, of an object detected or detectable by the distance sensor. Thus, as an embodiment, the echo density parameter of a first object having such a large area is smaller than the echo density parameter of the second object, even though it receives the same number of echoes from a second object having a comparatively smaller area.

[0030] The echo density parameter is advantageously determined as the ratio of the number of received echoes to a predetermined time measurement window and a predetermined area in the surrounding environment, and the number of received echoes is preferably normalized with respect to the duration of the measurement window and the predetermined area in the surrounding environment.

[0031] Instead of determining the echo density parameter as a ratio of the number of received echoes to a predetermined time measurement window, and / or as a ratio to a predetermined area in the surrounding environment, in a more advantageous embodiment, the echo density parameter is determined as a ratio of the number of received echoes to the number of emitted measurement signals, and the number of received echoes is preferably normalized to the number of emitted measurement signals.

[0032] In a more advantageous embodiment, each distance value is determined based on the received echo, the number of received echoes is weighted according to at least one distance value, and the echo density parameter is determined based on the weighted number of echoes. The number of received echoes is advantageously weighted in this case as a function of multiple distance values, in particular as a function of all the distance values ​​of the corresponding echoes. By taking distance into account and weighting in this way, the fact that the object detection behavior changes as a function of distance is taken into account. This can improve the reliability of height classification.

[0033] The distance sensor is advantageously designed as a radar sensor. However, in a preferred embodiment, the distance sensor is designed as an ultrasonic sensor, and the measurement signal is designed as an ultrasonic signal. In this regard, the ultrasonic sensor, in particular a 1D ultrasonic sensor, can be located, for example, inside or behind the bumper of a vehicle. Alternatively, the ultrasonic sensor, in particular a 1D ultrasonic sensor, can be located, for example, inside or behind a body component such as a vehicle door.

[0034] In a more advantageous embodiment, the method is applied during an auxiliary and / or semi-automatic and / or automatic parking procedure.

[0035] The present invention further comprises an assistance system having a distance sensor and a control device. In this specification, the control device is configured to carry out the method according to the present invention.

[0036] The advantages and preferred embodiments described with respect to the method according to the present invention also apply to the support system according to the present invention.

[0037] Illustrative embodiments of the present invention will be described in more detail below with reference to the drawings. [Brief explanation of the drawing]

[0038] [Figure 1] This shows a radiation diagram representing the radiation pattern of an ultrasonic sensor as a function of elevation angle. [Figure 2a] This shows the curve of the echo density parameter for low-level objects when a car is approaching. [Figure 2b] This shows the curve of the echo density parameter for high-objects when a car is approaching. [Figure 3] This diagram shows a flowchart of a method for characterizing objects in the surrounding environment of an automobile. [Modes for carrying out the invention]

[0039] Figure 1 shows a radiation diagram representing the radiation pattern 1 of an ultrasonic sensor in an automobile as a function of the elevation angle. From this, it can be seen that the radiation pattern 1 of the ultrasonic sensor is a function of the elevation angle, that is, the power of the ultrasonic signal emitted by the ultrasonic sensor to an object within the detection area depends on the elevation angle.

[0040] When an object is positioned at an elevation angle of 90°, that is, at least at the installation height of the ultrasonic sensor inside the vehicle, the elevation angle does not change as the vehicle, or more precisely the ultrasonic sensor, approaches the object. The power, i.e., amplitude, of the reflected ultrasonic signal or echo depends substantially on the distance between the ultrasonic sensor and the object. Therefore, as the vehicle or ultrasonic sensor approaches a tall object, the amplitude of the reflected ultrasonic signal gradually increases.

[0041] For objects that are lower in height than the installation height of the ultrasonic sensor inside the vehicle, the elevation angle, and therefore the power or amplitude of the reflected ultrasonic signal, will consequently vary depending on the distance between the vehicle or ultrasonic sensor and the object. This is because, as the vehicle or ultrasonic sensor approaches the object, the elevation angle decreases continuously until it reaches approximately 0° as soon as the ultrasonic sensor is directly positioned over the object. Therefore, as the distance decreases, i.e., as the vehicle continues to approach the object, the elevation angle decreases. This can lead to the amplitude of the reflected measurement signal decreasing to such an extent that the ultrasonic signal reflected by the object is not detected.

[0042] Figure 2a shows several curves 2, each representing the echo density parameter curve for low objects, such as curbs, as the vehicle approaches an object. Figure 2b, accordingly, shows several curves 3, each representing the echo density parameter curve for high objects, such as walls that are much taller than curbs, as the vehicle approaches an object. In this specification, the distance from the vehicle to the object is determined by a point on the horizontal coordinate system, and the value of the echo density parameter is determined by a point on the vertical coordinate system.

[0043] First, as is clearly visible in Figures 2a and 2b, when the distance from each object to the car or the car's ultrasonic sensor is approximately 2.4 meters, the echo density parameters for the curb and wall have similar values ​​in the range of approximately 0.8 to 0.9. As can also be seen from Figure 2a, starting with a distance of approximately 2 meters or slightly over 2 meters from the curb, as the car approaches the curb (further), the echo density parameter decreases significantly or sharply first to a distance of approximately 0.5 meters. If the car approaches the curb even further beyond that, the echo density parameter decreases further, in which case it approaches (ideally) a value of 0. In contrast, when the distance between the car or the car's ultrasonic sensor and the wall starts at approximately 2 meters or slightly over 2 meters, as the car approaches the wall (further), the echo density parameter increases significantly or sharply first to a distance of approximately 0.5 meters. If the car approaches the curb even further beyond that, the echo density parameter increases further, in which case it approaches (ideally) a value of 1.

[0044] As shown in Figure 2a, based on the fact that the echo density parameter curve drops when a vehicle approaches a curb, objects or curbs are classified as "low" in particular within a distance range of approximately 2 meters or slightly over 2 meters to 0.5 meters. On the other hand, as shown in Figure 2b, based on the fact that the echo density parameter curve rises when a vehicle approaches a wall, objects or walls are classified as "high" in particular within a distance range of approximately 2 meters or slightly over 2 meters to 0.5 meters.

[0045] Therefore, the height of an object can be classified inexpensively, simultaneously, accurately, and reliably based on the echo density parameter curve, in particular, without using the echo density parameter, its curve, or thresholds or limits for comparison with variables derived from the echo density parameter.

[0046] Figure 3 shows a flowchart of method 100 for characterizing an object in the surrounding environment of an automobile. The automobile in this regard is equipped with a support system having a control device and a 1D ultrasonic sensor having the radiation pattern shown in Figure 1. The automobile approaches the object in close proximity, for example, at a distance of about 2.5 meters, or from a distance of about 2.5 meters.

[0047] In step 101, an ultrasonic sensor continuously emits an ultrasonic signal, and an echo of the ultrasonic signal reflected by an object is received.

[0048] Step 102, which may be performed at least in part in parallel with Step 101, determines the number of echoes received by the object and an echo density parameter based on the number of received echoes, in particular using a sliding window, in a temporally continuous manner. According to the embodiment, the echo density parameter is defined as the ratio of the number of received echoes to the number of emitted ultrasonic signals.

[0049] In step 103, the height classification of the object is determined. For this purpose, the distance-dependent curve of the echo density parameter is evaluated, and the object is classified as either a tall object that a car cannot pass through, or a short object that a car can pass through, depending on whether the echo density parameter curve rises or falls.

[0050] In this way, the height of an object can be classified inexpensively and reliably without using echo density parameters, their curves, or thresholds or limits for comparison with variables derived from the echo density parameters.

Claims

1. A method (100) for characterizing an object in the surrounding environment of an automobile by an automobile assistance system, wherein the automobile is moved relative to the object, a distance sensor of the assistance system is used to emit a measurement signal, an echo of the measurement signal reflected by the object is received, and a control device of the assistance system is used to determine the height classification of the object based on a density parameter determined based on the number of echoes received. The method (100), characterized in that a curve of the echo density parameter is determined, and the classification of the height of the object is determined based on the curve of the echo density parameter.

2. The method according to claim 1 (100), characterized in that, in order to determine the classification of the height of the object, no comparison is made between the curve of the echo density parameter and, in particular, a predetermined threshold.

3. The method according to claim 1 or 2 (100), characterized in that the classification of the height of the object is determined based on a change in the curve of the echo density parameter, preferably an increase and / or decrease.

4. The method according to claim 3 (100), characterized in that the object is located near the automobile, preferably at a distance of up to 3 meters from the distance sensor of the automobile, the object is classified as "low" if a downward movement of the echo density parameter curve is determined when the automobile approaches the object, and the object is classified as "high" if an upward movement of the echo density parameter curve is determined.

5. The method according to any one of the prior claims (100), characterized in that the classification of the height of the object is determined based on the slope of the curve of the echo density parameter.

6. The method according to any one of the prior claims (100), characterized in that the echo density parameter is determined as a ratio of the number of received echoes to a predetermined time measurement window, and the number of received echoes is preferably normalized with respect to the duration of the measurement window.

7. The method according to any one of the prior claims (100), characterized in that the echo density parameter is determined as a ratio of the number of received echoes to a predetermined area in the surrounding environment, and the number of received echoes is preferably normalized with respect to the predetermined area in the surrounding environment.

8. The method according to any one of the prior claims (100), characterized in that the echo density parameter is determined as the ratio of the number of received echoes to the number of emitted measurement signals, and the number of received echoes is preferably normalized with respect to the number of emitted measurement signals.

9. The method according to any one of the prior claims (100), characterized in that each distance value is determined based on the received echo, the number of received echoes is weighted as a function of at least one distance value, and the echo density parameter is determined based on the weighted number of echoes.

10. The method according to any one of the prior claims (100), characterized in that the distance sensor is in the form of an ultrasonic sensor and the measurement signal is designed as an ultrasonic signal.

11. The method according to any one of the prior claims (100), characterized in that the method is applied during an auxiliary and / or semi-automatic and / or automatic parking procedure.

12. A support system comprising a distance sensor and a control device designed to perform the method (100) described in any one of the prior claims.

Images