EP4314895A1

EP4314895A1 - Method for operating a driver assistance system, computer program product, driver assistance system, and vehicle

Info

Publication number: EP4314895A1
Application number: EP22718982.6A
Authority: EP
Inventors: Ludovic Mosnier-Thoumas; Markus Heimberger; Niko Moritz Scholz; Jean-Francois Bariant
Original assignee: Valeo Schalter und Sensoren GmbH
Current assignee: Valeo Schalter und Sensoren GmbH
Priority date: 2021-03-30
Filing date: 2022-03-30
Publication date: 2024-02-07
Also published as: JP2024511664A; WO2022207679A1; CN117157555A; KR20230164145A; DE102021107972A1; US20240194077A1

Abstract

The invention relates to a method for operating a driver assistance system (110). The method has the steps of: a) receiving (S1) a drive state sensor signal (SIG0(t)), which indicates the drive state, at a number of different points in time (t0 - t5), b) receiving (S2) a number of sensor signals (SIG1(t)), which indicate the surroundings (200), at a number of different points in time (t0 - t5), c) detecting (S3) a number of objects (210, 211) in the surroundings (200) on the basis of a first number of sensor signals (SIG1(t)), which have been detected at a first point in time, d) ascertaining (S4) a position (POS) and a movement vector (VEC) for a detected object (210, 211) on the basis of the first number of sensor signals (SIG1(t)) and a second number of sensor signals (SIG1(t)), which have been received at a second point in time following the first point in time, using a plurality of different ascertaining methods (V1, V2), wherein different ascertaining methods (V1, V2) of the plurality have a different degree of computing complexity, and e) outputting (S5) a warning signal if a potential collision with the detected object (210, 211) is ascertained on the basis of the drive state sensor signal (SIG0(t)) received at a specified point in time and the position (POS) and the movement vector (VEC) ascertained for the detected object (210, 211).

Description

METHOD OF OPERATING A DRIVING ASSISTANCE SYSTEM, COMPUTER PROGRAM PRODUCT, DRIVING ASSISTANCE SYSTEM AND VEHICLE

The present invention relates to a method for operating a driver assistance system, a computer program product, a driver assistance system and a vehicle with such a driver assistance system.

Known vehicles have a number of sensor units with which they can detect their surroundings, such as ultrasonic sensors. This can be particularly helpful when maneuvering and/or parking the vehicle, especially if the vehicle is large and/or difficult to see. Based on the sensor data, an acoustic, haptic and/or visual signal can be output to the driver to warn him of a collision. One difficulty is detecting moving (dynamic) objects and determining whether a collision with them can occur. A dynamic object is, in particular, another road user, such as a pedestrian or a cyclist.

DE 102006045418 A1 discloses a motor vehicle with a driver assistance system and with a sensor for measuring the distance from an obstacle. It is proposed to detect a movement direction of a moving object in order to increase road safety.

Known methods for determining dynamic objects are computationally intensive and complex, which is why error detection can occur more often and/or the time from receiving the sensor signal to the result is quite long. This reduces the time left to react if a collision is imminent. Furthermore, known systems require expensive sensors, such as radar, a lidar or a camera. Against this background, an object of the present invention is to improve the operation of a driver assistance system.

According to a first aspect, a method for operating a driver assistance system for a vehicle is proposed. The method includes: a) receiving a driving condition sensor signal indicative of a driving condition of the vehicle at a number of different points in time, b) receiving a number of sensor signals indicative of an area surrounding the vehicle at a number of different points in time, c) detecting a number of objects in the area of the vehicle as a function of a first number of sensor signals received at a first point in time, d) determining a position and a motion vector for a detected object as a function of the first number of sensor signals and a second number of sensor signals which are received at a second , the time following the first time was received using a plurality of different determination methods, wherein different determination methods of the plurality have a different computing effort, and e) outputting a warning signal when a potential collision of the vehicle with de m detected object is determined on the basis of the driving condition sensor signal received at a specific point in time and the position determined for the detected object and its movement vector.

This method has the advantage that the movement of detected objects is determined using different determination methods. In particular, a less complex determination method that can be carried out very quickly and a more complex determination method that can be carried out somewhat more slowly can be used here. The different determination methods have, for example, different levels of accuracy and/or reliability with regard to their respective results. So in particular a less complex investigative procedures can be less reliable, but save time, especially in critical situations.

The driving condition sensor signal includes, for example, odometry data of the vehicle, such as a current speed, a current wheel speed, a current wheel angle and/or a current steering angle. In particular, a direction of the vehicle, for example a future trajectory or a driving path, can be determined on the basis of the driving state sensor signal.

In the present case, the fact that the driving condition sensor signal is received at a number of different points in time is to be understood in particular to mean that the driving condition sensor signal that is current at this point in time is received at a particular point in time. The current driving condition sensor signal is indicative in particular of the current driving condition at the current point in time. For example, the driving condition sensor signal is received regularly, in particular periodically, for example with a frequency of more than 1 Hz, preferably at least 10 Hz, preferably up to 100 Hz. Based on at least consecutive driving condition sensor signals, a change in the driving condition can be determined, for example.

The sensor signals indicative of the surroundings of the vehicle include, in particular, ultrasonic sensor signals. In the present case, the fact that a number of sensor signals are received is to be understood in particular as meaning, for example, that sensor signals are received from a number of different sensors, with different sensors on the one hand comprising sensors of the same type but with a different arrangement and/or orientation, but on the other hand also comprising sensors of different types , such as an ultrasonic sensor and a camera. In this case, the number comprises a quantity greater than or equal to one. The sensor signals can still be received from virtual sensors. Alternatively, one can also say that the sensor signals are retrieved. For example, a sensor provides an output signal that can be called up by external components, such as the driver assistance system. In the present case, the fact that the number of sensor signals are received at a number of different points in time is to be understood in particular to mean that the sensor signals that are current at this point in time are received at a particular point in time. All of the sensor signals of the number are preferably received at one point in time, for example as a data packet that includes all of the sensor signals of the number.

It should be noted that different sensor signals of the number, for example sensor signals from two different ultrasonic sensors, can have different detection times, even if the number of sensor signals is received at one time.

It can also be the case that an interval between two sensor signals from a respective sensor is different from an interval between two sensor signals from a further sensor. Each sensor preferably provides its sensor signal regularly, in particular periodically, for example with a frequency of more than 1 Hz, preferably at least 10 Hz, preferably up to 100 Hz, with the current sensor signal always being provided at a particular point in time.

The times at which the number of sensor signals is received may differ from the times at which the driving condition sensor signal is received.

In the present case, the fact that a number of objects in the area surrounding the vehicle are detected as a function of a first number of sensor signals that were received at a first point in time means in particular that the first number of received sensor signals is processed and/or analyzed , and as a result of the processing and/or the analyzing a number of objects are detected. The processing and/or analysis includes, for example, a signal analysis of the individual sensor signals and/or a signal analysis of a number of sensor signals that are correlated with one another. The number of objects is greater than or equal to one.

A position and a motion vector are determined for at least one of the detected objects. A single sensor signal at a point in time can be sufficient to determine the position. At least two sensor signals, which were received at different points in time, are necessary to determine the movement vector. Two sensor signals of a specific sensor received at different times are preferably used to determine the movement vector, but sensor signals from different sensors that were received at different times can also be used. The more sensor signals are used for a respective object, the more accurate the determination of the position and the motion vector can be. In addition, it is advantageous to keep the time interval between two sensor signals as short as possible in order to improve accuracy and timeliness, since the movement vector can change at any time, for example if a pedestrian stops abruptly or starts moving.

The position of the object relates in particular to a position in a coordinate system of the vehicle and has at least two coordinates. The position of the object can refer to a single point of the object.

In particular, the motion vector comprises a two-dimensional vector. A magnitude of the motion vector corresponds, for example, to the speed of the object.

The plurality of different determination methods includes at least two determination methods that have different computational complexity. For example, in a first determination method, the position is determined and the motion vector is determined directly on the basis of raw sensor data, and in a second determination method, processed data is generated on the basis of the raw data and the position is determined and the motion vector is determined on the basis of the processed data. Sensor raw data are in particular the unprocessed output signal of a respective sensors. Since there is no pre-processing, this determination method can be particularly fast. On the other hand, signal noise or the like may affect a determination result.

The fewer processing steps a discovery method has, the less complex it is and the faster it can be performed. The plurality of different determination methods therefore advantageously includes at least two determination methods with a different number of processing steps. A processing step includes, for example, performing a particular mathematical operation on a respective sensor signal, such as forming a moving average to mask outliers in the sensor signal, or applying a noise filter, a Fourier transform, and the like. A processing step can also include a number of sensor signals, for example when a correlation is determined and/or the sensor signals are mutually checked for plausibility.

Different positions and/or movement vectors can be determined for an object on the basis of the various determination methods. The most up-to-date position and the most up-to-date movement vector are preferably always used as the basis for determining a possible collision. If a position and a motion vector for an object were determined by two different determination methods, which are based on the same current sensor signals, then in particular that position and that motion vector are used as the basis for determining a possible collision, their respective accuracy and/or reliability is higher. The accuracy and/or reliability can be determined, for example, in the form of a determination error, which can originate on the one hand from measurement errors and on the other hand from the determination method itself.

Preferably, the position and the motion vector are determined for several of the detected objects of the number, preferably for each detected object of the number. In particular, the position and the motion vector are constantly updated. For example, a new determination is made whenever a current sensor signal or a number of current sensor signals is received. The current position and the current movement vector can therefore also be used in the following.

Based on the driving condition sensor signal received at a specific point in time and the position determined for the detected object and its movement vector, it can be determined whether the vehicle may collide with the object moving according to its movement vector in the future. Here, in particular, an extrapolation of the position of the vehicle and the object based on the respective current driving state sensor signal and an extrapolation of the position of the object based on the current position and the current movement vector of the object can be determined.

For example, if the extrapolated positions get too close at a certain point in time, a collision is likely.

A warning signal is given when a collision is likely. The warning signal can be output directly to the user of the vehicle, for example as an acoustic, haptic and/or visual signal. The warning signal can also be output to the outside of the vehicle, for example to the object, in particular as an acoustic warning signal. If the warning signal is issued, further functions of the driver assistance system and/or other units of the vehicle can also be triggered.

The warning signal is preferably output independently of the determination method used to determine the position and the movement vector of the object with which a possible collision was determined. According to one embodiment of the method, the number of different determination methods includes at least one first determination method in which for each detected object of the number a Kalman filter is assigned and initialized, which is used to determine the position and the motion vector of the respective object.

This means that with five objects detected, five associated Kalman filters are initialized. The Kalman filter (also Kalman-Bucy filter, Stratonovich-Kalman-Bucy filter or Kalman-Bucy-Stratonovich filter) is a mathematical method for the iterative estimation of parameters, in this case the position and the motion vector of the object based on erroneous measurements, in this case the received sensor signals. The Kalman filter is used to estimate system variables that cannot be measured directly, while the errors in the measurements are optimally reduced. The Kalman filter describes an estimated value using multidimensional normal distributions. These represent a probability distribution of possible errors around each estimate, as well as correlations between estimate errors of different variables. With this information, the previous estimated values are optimally combined with the new measurements in each time step, so that remaining errors are minimized as quickly as possible. At a current point in time, the Kalman filter has a filter state comprising the current estimated values as well as error estimates and correlations. After each new measurement, the Kalman filter improves the previous estimates and updates the associated error estimates and correlations. In dynamic systems in which, for example, a speed is also estimated, the Kalman filter also estimates correlations between the speed and, for example, the position, in particular on the basis of equations of motion, and takes these into account for the next time step.

The Kalman filter is preferably updated at least whenever a number of current sensor signals have been received. According to one embodiment of the method, different sensor signals of the number of different scanning areas in the environment are assigned, with a respective sensor signal of the number of sensor signals received at a particular point in time, assigned to a specific scanning area in the area, being fed to that Kalman filter whose assigned object has a position , which lies within the scanning range assigned to the sensor signal.

This ensures that the respective Kalman filter is supplied with that sensor signal for the next update which also relates to the object assigned to the Kalman filter.

According to a further embodiment of the method, the warning signal is output if a potential collision is determined on the basis of the position determined for the respective detected object using the first determination method and its motion vector, only if the determined motion vector of the object is not equal to zero.

In other words, there is no warning of a possible collision with an object if the possible collision was determined based on the position determined using the first determination method and, for example, a movement of the vehicle, but the object itself is not moving but is static. It should be noted that this does not preclude warnings of such a collision. For example, the possible collision can also be determined on the basis of the position of the object determined using a further determination method, and a warning can then be issued accordingly.

According to a further embodiment of the method, this includes:

Determining a driving path for the vehicle depending on the received driving condition sensor signal. In the present case, the term “travel path” describes in particular the area that the vehicle would sweep over if it were moved forwards or backwards with the current wheel angle or steering angle. This means that a change in the steering angle or the wheel angle causes a change in the driving path. The driving path can be represented, for example, by trajectories for each wheel of the vehicle. The driving path can also be understood as a two-dimensional future trajectory of the vehicle.

According to a further embodiment of the method, a warning signal is only output when a distance of the respective object from the vehicle and/or from the determined driving path is less than or equal to a lower threshold value.

The threshold value can be variably determined in particular for different vehicles and/or situations. Furthermore, the threshold value can be determined as a function of the vehicle speed and/or the speed of the object. Furthermore, the threshold value can be determined as a function of a measurement accuracy, such as a standard deviation or a variance.

This embodiment has the advantage that a warning is only issued when the probability of a collision is relatively high. In particular, the threshold value takes into account that it is unknown how the object will continue to move, ie whether it will change its speed and/or direction, for example. The threshold can, for example, assume a value from an interval from zero to two meters.

According to a further embodiment of the method, a warning signal is only output when the determined movement vector of the respective object points in the direction of the vehicle and/or the determined driving path.

This has the advantage that, for example, in the case of an object whose current distance is below the lower threshold value, but which is away from the vehicle or the driving path moved away and with which a collision is therefore very unlikely, no warning signal is issued.

According to a further embodiment of the method, step e) comprises:

Determining a future trajectory of the detected object on the basis of the determined position and the movement vector, with a warning signal only being output if the determined future trajectory falls below a predetermined minimum distance at at least one position and/or has an intersection with the determined driving path.

The future trajectory can be determined, for example, as an extrapolation of a previous trajectory of the object. In particular, a curved future trajectory can also be determined here.

The predetermined minimum distance can be equal to the lower threshold value, but it can also be different from it. The predetermined minimum distance can be variably determined, in particular for different vehicles and/or situations. Furthermore, the predetermined minimum distance can be determined as a function of the vehicle speed and/or the speed of the object. Furthermore, the predetermined minimum distance can be determined as a function of a measurement accuracy, such as a standard deviation or a variance.

The fact that the predetermined minimum distance is determined as a function of another variable, such as the vehicle speed, means, for example, that the predetermined minimum distance is predetermined as a function of the other variable, so that a specific numerical value at a point in time depends on the current value of the other quantity is determined.

According to a further embodiment of the method, the received sensor signals exclusively include ultrasonic sensor signals. This embodiment can be used advantageously for vehicles that only have ultrasonic sensors. Compared to other sensors, such as radar, lidar and/or cameras, ultrasonic sensors are inexpensive to manufacture and do not require high computing power to evaluate their sensor signals.

This embodiment can also be advantageous for vehicles that have additional sensors, since the computing power required for the proposed method is less than for alternative methods that, for example, additionally carry out an image evaluation of a camera image.

According to a further embodiment of the method, the number of different determination methods includes at least one second determination method in which a feature recognition is carried out on the basis of the number of sensor signals received at a particular point in time and a digital map of the surroundings is determined using recognized features.

The digital map of the surroundings can advantageously be used to also determine a collision with static objects and to warn of this. In comparison to the first determination method, carrying out a feature detection (English: "feature extraction") is complex and requires more computing power. Therefore, the second preliminary investigation may last longer. Furthermore, it can happen that moving objects are not detected during feature detection, for example if the sensor signal changes quickly and therefore looks like noise. This can be the case in particular with pedestrians and/or cyclists.

According to a further embodiment of the method, the method is only carried out when the vehicle has a speed of less than or equal to 15 km/h, preferably less than or equal to 10 km/h, preferably less than or equal to 7 km/h, more preferably less than or equal to 5 km/h. The vehicle speed can be determined in particular on the basis of the driving condition sensor signal.

This is particularly advantageous when using ultrasonic sensor signals, since ultrasonic sensors work comparatively slowly due to the propagation speed of sound in air and are therefore less suitable for high speeds.

According to a second aspect, a computer program product is proposed which comprises instructions which, when the program is executed by a computer, cause the latter to execute the method according to the first aspect.

A computer program product, such as a computer program means, can be made available or supplied by a server in a network, for example, as a storage medium such as a memory card, USB stick, CD-ROM, DVD, or in the form of a downloadable file. This can be done, for example, in a wireless communication network by transferring a corresponding file with the computer program product or the computer program means.

According to a third aspect, a driver assistance system for a vehicle is proposed. The driver assistance system comprises a receiving unit for receiving a driving condition sensor signal indicative of a driving condition of the vehicle at a number of different points in time, and for receiving a number of sensor signals indicative of an area surrounding the vehicle at a number of different points in time, a detection unit for detecting a number of objects in the Environment of the vehicle depending on a first number of sensor signals received at a first point in time, a determination unit for determining a position and a movement vector for a detected object of the number depending on the first number of sensor signals and a second number of sensor signals, received at a second time subsequent to the first time using a plurality different determination methods, wherein different determination methods of the plurality have a different computing effort, and an output unit for outputting a warning signal if there is a potential collision of the vehicle with the detected object on the basis of the driving condition sensor signal received at a specific point in time and the position determined for the detected object and its Motion vector is determined.

The embodiments and features described for the proposed method apply accordingly to the proposed driver assistance system. The advantages and/or definitions that were mentioned in relation to the method according to the first aspect also apply to the proposed driver assistance system. The driver assistance system is operated in particular with the method according to the first aspect or one of the specific embodiments of the method.

The respective unit of the driver assistance system can be implemented in terms of hardware and/or software. In the case of a hardware implementation, the respective unit can be embodied, for example, as a computer or as a microprocessor. In the case of a software implementation, the respective unit can be designed as a computer program product, as a function, as a routine, as an algorithm, as part of a program code or as an executable object. Furthermore, each of the units mentioned here can also be designed as part of a higher-level control system of the vehicle, such as a central electronic control device and/or an engine control unit (ECU: Engine Control Unit).

The driver assistance system can be set up in particular for semi-autonomous or fully autonomous driving of the vehicle. Partially autonomous driving is understood to mean, for example, that the driver assistance system controls a steering device and/or an automatic driving stage. Fully autonomous driving means, for example, that the driver assistance system also controls a drive device and a braking device. According to a fourth aspect, a vehicle with a number of surroundings sensor units for detecting surroundings of the vehicle and for outputting a respective sensor signal and with a driver assistance system according to the third aspect.

The vehicle is, for example, a passenger car or a truck. The vehicle preferably includes a number of sensor units that are set up to detect the driving state of the vehicle and to detect an environment of the vehicle. Examples of such sensor units of the vehicle are imaging devices, such as a camera, radar (radio detection and ranging) or a lidar (engl, light detection and ranging), ultrasonic sensors, location sensors, wheel angle sensors and/or wheel speed sensors. The sensor units are each set up to output a sensor signal, for example to the driver assistance system, which carries out the partially autonomous or fully autonomous driving as a function of the detected sensor signals.

According to one embodiment of the vehicle, the environmental sensor units exclusively include ultrasonic sensors.

This allows for a vehicle that is less complex and less expensive to stow.

According to a further embodiment of the vehicle, the vehicle has a mass of more than 2.5 tons and/or a length of more than 5 meters.

For example, the vehicle is designed as a transporter. Vans, for example, are confusing and can have one or more "blind spots" that a driver of the van can see poorly or not at all. Such a blind spot is, for example, on the passenger side behind the A-pillar or on the driver and/or passenger side close to the vehicle. Particularly in the case of large and/or unclear vehicles, the method has the advantage that a warning can be given in the event of an imminent collision with an object that is in an area close to the vehicle that is difficult or impossible for the driver to see.

Further possible implementations of the invention also include combinations of features or embodiments described above or below with regard to the exemplary embodiments that are not explicitly mentioned. The person skilled in the art will also add individual aspects as improvements or additions to the respective basic form of the invention.

Further advantageous refinements and aspects of the invention are the subject matter of the dependent claims and of the exemplary embodiments of the invention described below. The invention is explained in more detail below on the basis of preferred embodiments with reference to the enclosed figures.

1 shows a schematic view of an exemplary embodiment of a vehicle;

Fig. 2 shows a schematic view of different scanning areas;

3 shows a schematic view of a first traffic situation;

4 shows a schematic view of a second traffic situation;

5 shows a schematic view of a third traffic situation at different points in time;

6 shows a schematic block diagram of an exemplary embodiment of a driver assistance system; and FIG. 7 shows a schematic block diagram of an exemplary embodiment of a method for operating a driver assistance system.

Elements that are the same or have the same function have been provided with the same reference symbols in the figures, unless otherwise stated.

1 shows a schematic view of a vehicle 100 from a bird's eye view.

The vehicle 100 is, for example, a car that is arranged in an environment 200 . Car 100 has a driver assistance system 110, which is embodied as a control unit, for example. In addition, a plurality of environment sensor devices 120, 130 are arranged on car 100, these being optical sensors 120 and ultrasonic sensors 130, for example. The optical sensors 120 include, for example, visual cameras, a radar and/or a lidar. The optical sensors 120 can each capture an image of a respective area from the environment 200 of the car 100 and output it as an optical sensor signal. The ultrasonic sensors 130 are set up in particular for scanning a respective area 131 - 136 (see FIG. 2 ) of the environment 200 . On the basis of the sensor signal output by the ultrasonic sensors 130, objects 210, 211 (see FIGS.

211 determine. For example, a motion vector VEC (see FIGS. 3-5) of an object 210, 211 can be determined from sensor signals that follow one another in time. Using the sensor signals detected by sensors 120, 130, driver assistance system 110 may be able to drive car 100 partially or fully autonomously. In addition to the optical sensors 120 and ultrasonic sensors 130 shown in FIG. 1, it can be provided that the vehicle 100 has various other sensor devices 120, 130. Examples of this are a microphone, an acceleration sensor, a wheel speed sensor, a wheel angle sensor, a steering angle sensor, an antenna with a coupled receiver for receiving electromagnetically transmittable data signals, and the like. Driver assistance system 110 is designed, for example, as explained with reference to FIG. 6 and set up to carry out the method described with reference to FIG. 7 . Driver assistance system 110 is preferably also set up to carry out processing procedures as described below with reference to FIGS. 2-5.

2 shows a schematic view of different scanning areas 131-136 of different ultrasonic sensors 130. In this example, six ultrasonic sensors 130 are arranged on a front sill of a vehicle 100. Vehicle 100 is designed, for example, as described with reference to FIG. 1 . Each sensor 130 has a specific scanning range 131-136. The shape of a respective scanning area 131 - 136 depends on the arrangement and alignment of ultrasonic sensor 130 on vehicle 100, but also on the structure of ultrasonic sensor 130. The scanning areas 131 - 136 can at least partially overlap, so that the immediate surroundings 200 in front of the sill of vehicle 100 can preferably be detected without any gaps. The range of a respective ultrasonic sensor 130 depends on its structure, for example in a range between five meters and ten meters.

In addition to the six ultrasonic sensors 130 shown here that are physically present, additional virtual ultrasonic sensors (not shown) may be present. A virtual ultrasonic sensor is based, for example, on a first ultrasonic sensor 130 emitting an ultrasonic signal and a second ultrasonic sensor 130 receiving a reflection of the ultrasonic signal emitted by the first ultrasonic sensor. A virtual ultrasonic sensor has, for example, a virtual position between two ultrasonic sensors 130 that are physically present.

There are two objects 210, 211 in the surroundings 200 of the vehicle 100. A first object 210, for example a cyclist, is located in the scanning areas 135, 136 of two ultrasonic sensors 130. The cyclist 210 is therefore detected by two ultrasonic sensors in particular. In addition, the cyclist can be detected by a virtual ultrasonic sensor as described above. A second object 211, for example a Pedestrian is located in the scanning area 132 of a single ultrasonic sensor 130. However, the pedestrian 211 can also be detected by a virtual ultrasonic sensor as described above.

To determine the position POS (see Fig. 7) and the motion vector VEC (see Fig. 3 - 5) of a respective object 210, 211 by means of a first determination method V1 (see Fig. 6 or 7), each detected object 210, 211 is assigned a Kalman filter mapped and initialized. In this example, therefore, two Kalman filters are initialized. A respective Kalman filter is set up to estimate the position of the respective object 210, 211 on the basis of the ultrasonic sensor signals SIG1(t) received one after the other. In this case, the position includes in particular the position POS and the movement vector VEC of the respective object 210, 211. In particular, each Kalman filter is supplied with the ultrasonic sensor signals SIG1(t) received from those ultrasonic sensors 130 in whose scanning range 131 - 136 the respective object 210, 211 is located just located. A precise and consistent result and an exact tracking of the objects 210, 211 are thus possible. A second determination method V2 (see FIG. 6 or 7) can provide that on the basis of the number of sensor signals SIG1(t) received at a particular point in time tO-15 (see FIG. 5) a feature recognition is carried out and a digital environment map is used is determined from recognized features.

FIG. 3 shows a schematic view of a first traffic situation, in which, for example, vehicle 100 from FIG. 1 or FIG. 2 is shown on a road. An object 210 , for example a pedestrian, is shown on the right in front of vehicle 100 . The driving path TR for the vehicle 100 is also shown. Driving path TR is determined, for example, by driver assistance system 110 (see FIG. 1 or 6) on the basis of a driving condition sensor signal SIG0(t) (see FIG. 6 or 7), which includes a current steering angle or a current wheel angle. The ultrasonic sensors 130 (see FIG. 1 or 2) preferably constantly emit ultrasonic signals and detect the reflected signals, ie they constantly scan their respective scanning area 131-136 (see FIG. 2) with ultrasonic signals. For example, the sampling occurs 10 times per second, preferably at least 50 times per second, preferably at least 100 times per second. Ultrasonic sensors 130 emit ultrasonic sensor signals SIG1(t) (see FIG. 6 or 7) with a corresponding frequency, for example to driver assistance system 110. Based on the ultrasonic sensor signals, a position POS (see FIG. 7) of pedestrian 210 can be inferred. A movement vector VEC for the pedestrian 210 can also be determined on the basis of at least two ultrasonic sensor signals SIG(t) recorded one after the other. This is done, for example, as described with reference to FIG. 2 using a first determination method V1

In the situation shown, pedestrian 210 is moving toward route path TR of vehicle 100 . The current distance D of the pedestrian 210 from the driving path TR is also shown. Driver assistance system 110 is set up to output a warning signal as a function of predetermined criteria. For example, it is checked whether the distance D of the pedestrian 210 from the current driving path TR (alternatively from the vehicle 100) is less than or equal to a predetermined threshold value, or whether the determined movement vector VEC points in the direction of the driving path TR or the vehicle 100. If one or more of these criteria are met, then the warning signal is issued since a collision with the pedestrian 210 is then likely unless the vehicle 100 is stopped or changes direction.

In other words, the warning signal is issued when there is a potential collision of the vehicle 100 with the detected object 210, 211 based on the driving condition sensor signal SIG0(t) received at a specific point in time tO - 15 (see FIG. 5) and that for the detected object 210, 211 determined position POS and its motion vector VEC is determined. FIG. 4 shows a schematic view of a second traffic situation, in which, for example, vehicle 100 from FIG. 1 or FIG. 2 is shown on a road. An object 210 , for example a pedestrian, is shown on the right in front of vehicle 100 . The driving path TR for the vehicle 100 is also shown. Driving path TR is determined, for example, by driver assistance system 110 (see FIG. 1 or 6) on the basis of a driving condition sensor signal SIG0(t) (see FIG. 6 or 7), which includes a current steering angle or a current wheel angle.

A position POS (see FIG. 7) and a movement vector VEC of the pedestrian 210 are determined on the basis of ultrasonic sensor signals SIG1(t) (see FIG. 6 or 7). In addition, a future trajectory TR1 of pedestrian 210 is determined in this example. For example, the previous trajectory of the pedestrian 210 is extrapolated for this purpose. For example, the future trajectory TR1 can be determined on the basis of a specific embodiment of the first determination method V1, ie using Kalman filters. Additionally and/or alternatively, the future trajectory TR1 can be determined on the basis of a third determination method.

A smallest distance between the driving path TR and the future trajectory TR1 can be determined. If this distance D is less than a predetermined minimum distance, a warning signal is output, for example.

FIG. 5 shows a schematic view of a third traffic situation at different points in time tO−15, vehicle 100 of FIG. 1 or FIG. 2 being shown on a road, for example. An object 210 is determined to the right of vehicle 100 at an initial time tO. This takes place in particular on the basis of a number of sensor signals SIG1(t) received at the start time t0 (see FIG. 6 or 7). At a subsequent first point in time t1, a second number of sensor signals SIG1(t) are received. A current position POS (see FIG. 7) of the object 210(t1) is determined on the basis of the second number of sensor signals. Furthermore, based on the first number and the second number of sensor signals SIG1(t), a current motion vector VEC(t1) for Time t1 are determined. At a subsequent second point in time t2, a third number of sensor signals SIG1(t) are received and a current position POS of object 210(t2) at point in time t2 and a current motion vector VEC(t2) at point in time t2 are determined. At a subsequent third time t3, a fourth number of sensor signals SIG1(t) are received and a current position POS of the object 210(t3) at time t3 and a current motion vector VEC(t3) at time t3 are determined. At a subsequent fourth time t4, a fifth number of sensor signals SIG1(t) are received and a current position POS of the object 210(t4) at time t4 and a current motion vector VEC(t4) at time t4 are determined. At a subsequent fifth time t5, a sixth number of sensor signals SIG1(t) are received and a current position POS of the object 210(t5) at time t5 and a motion vector VEC(t5) current at time t5 are determined. The movement of the object 210 can thus be tracked at any point in time t0−15. In embodiments, the movement of the object 210 can additionally be predicted, for example using corresponding equations of motion.

The determination of the position POS and the motion vector VEC at a respective point in time t0 - 15 is preferably based on the first determination method V1 using a Kalman filter and on the basis of a further determination method V2 (see Fig. 6 or 7)

Fig. 6 shows a schematic block diagram of an exemplary embodiment of a driver assistance system 110, for example driver assistance system 110 of vehicle 100 in Fig.

1. The driving assistance system 110 includes a receiving unit 112 for receiving a driving condition sensor signal SIG0(t) indicative of a driving condition of the vehicle 100 at a number of different points in time tO-15 (see FIG. 5), and for receiving a number of signals for an environment 200 ( see Fig. 1 or 2) of the vehicle 100 indicative sensor signals SIG1 (t) at a number of different times tO - 15. The driver assistance system 110 further includes a detection unit 114 for detecting a number of objects 210, 211 (see Fig. 2) in the Environment 200 of the vehicle 100 depending from a first number of sensor signals SIG1(t), which were received at a first point in time, a determination unit 116 for determining a position POS (see Fig. 7) and a motion vector VEC (see Fig. 3 - 5) for a detected object 210 , 211 as a function of the first number of sensor signals SIG(t) and a second number of sensor signals SIG(t), which was received at a second point in time following the first point in time, using a plurality of different determination methods V1, V2, where different determination methods of the plurality have a different computing effort, and an output unit 118 for outputting a warning signal if there is a potential collision of vehicle 100 with the detected object 210, 211 on the basis of the driving condition sensor signal SIG0(t) received at a specific point in time tO - 15 and the position POS determined for the detected object 210, 211 and its motion vector VEC.

Fig. 7 shows a schematic block diagram of an exemplary embodiment of a method for operating a driver assistance system 110, for example the driver assistance system 110 of Fig. 6 or the driver assistance system 110 of the vehicle 100 of Fig. 1. In a first step S1, a for a driving state of the vehicle 100 indicative vehicle condition sensor signal SIG0(t) at a number of different times tO-15 (see Figure 5). In a second step S2, a number of sensor signals SIG1(t) indicative of an area 200 (see FIG. 1 or 2) of vehicle 100 is received at a number of different points in time t0-15. In a third step S3, a number of objects 210, 211 (see FIG. 2) in the surroundings 200 of vehicle 100 are detected as a function of a first number of sensor signals SIG1(t), which were received at a first point in time. In a fourth step S4, a position POS and a movement vector VEC (see FIGS. 3-5) for a detected object 210, 211 are determined as a function of the first number of sensor signals SIG1(t) and a second number of sensor signals SIG1(t) , which was received at a second point in time following the first point in time, determined using a plurality of different determination methods V1, V2, with different determination methods V1, V2 of the plurality having a different chen calculation effort. In a fifth step S5, a warning signal is output if there is a potential collision of the vehicle 100 with the detected object 110 based on the driving condition sensor signal SIG0(t) received at a specific point in time tO - 15 and the position POS determined for the detected object 210 and its Movement vector VEC is determined.

Although the present invention has been described using exemplary embodiments, it can be modified in many ways.

REFERENCE LIST

100 vehicle 110 driver assistance system 112 receiving unit 114 detection unit 116 determination unit 118 output unit 120 sensor

130 sensors

131 scan area

132 scan area

133 scanning area

134 scan area

135 scan area

136 scan area 200 environment

210 object 210(t0) object 210(t1) object 210(t2) object 210(t3) object 210(t4) object 210(t5) object

211 object

D distance POS position S1 process step

S2 method step S3 method step

54 process step

55 Process step SIGO(t) Driving status sensor signal SIG1(t) Sensor signal t Time t0 Time t1 Time t2 Time t3 Time t4 Time t5 Time TR driving path TR1 trajectory V1 determination method V2 determination method VEC motion vector VEC(t1) motion vector VEC(t2) motion vector VEC(t3) Motion Vector VEC(t4) Motion Vector VEC(t5) Motion Vector

Claims

PATENT CLAIMS

1. A method for operating a driver assistance system (110) for a vehicle (100), the method comprising: a) receiving (S1) a driving state sensor signal (SIGO(t)) indicative of a driving state of the vehicle (100) at a number of different points in time ( tO - 15), b) receiving (S2) a number of sensor signals (SIG1(t)) indicative of an environment (200) of the vehicle (100) at a number of different points in time (tO - 15), c) detecting (S3) a number of objects (210, 211) in the area (200) of the vehicle (100) as a function of a first number of sensor signals (SIG1 (t)) received at a first point in time, d) determining (S4) a Position (POS) and a motion vector (VEC) for a detected object (210, 211) as a function of the first number of sensor signals (SIG1 (t)) and a second number of sensor signals (SIG1 (t)) resulting in a second , time point following the first time point was received using ei ner plurality of different determination methods (V1, V2), wherein different determination methods (V1, V2) of the plurality have a different computing effort, and e) outputting (S5) of a warning signal if there is a potential collision of the vehicle (100) with the detected object (210 , 211) on the basis of the driving condition sensor signal (SIG0(t)) received at a specific point in time and the position (POS) determined for the detected object (210, 211) and its motion vector (VEC).

2. The method as claimed in claim 1, characterized in that the number of different determination methods (V1, V2) comprises at least one first determination method in which for each detected object (210, 211) the number is assigned and initialized a Kalman filter which, for Determining the position (POS) and the motion vector (VEC) of the respective object (210, 211) is used.

3. The method according to claim 2, characterized in that different sensor signals (SIG1 (t)) of the number of different scanning areas (131 - 136) in the area (200), wherein a sensor signal (SIG1(t)) of the number of sensor signals (SIG1(t)) received at a particular point in time and assigned to a specific scanning area (131-136) in the area (200) is assigned to that Kalman filter is supplied, the associated object (210, 211) of which has a position (POS) which lies within the scanning range (131 - 136) associated with the sensor signal (SIG1 (t)).

4. The method according to claim 2 or 3, characterized in that the output of the warning signal when a potential collision based on the for the respective detected object (210, 211) using the first determination method (V1) determined position (POS) and its Motion vector (VEC) is determined, only takes place when the determined motion vector (VEC) of the object (210, 211) is not equal to zero.

5. The method according to any one of the preceding claims, characterized by: determining a driving path (TR) for the vehicle (100) as a function of the received driving condition sensor signal (SIG0(t)).

6. The method according to claim 5, characterized in that a warning signal is only output when a distance (D) of the respective object (210, 211) to the vehicle (100) and / or to the determined driving path (TR) is smaller or is equal to a lower threshold.

7. The method according to claim 5 or 6, characterized in that a warning signal is only output if the determined movement vector (VEC) of the respective object (210, 211) in the direction of the vehicle (100) and/or the determined driving path ( TR) points.

8. The method according to any one of claims 5 to 7, characterized in that step e) comprises:

Determining a future trajectory (TR1) of the detected object (210, 211) based on the determined position (POS) and the motion vector (VEC), with a warning signal is only output if the determined future trajectory (TR1) falls below a predetermined minimum distance at at least one position and/or has an intersection with the determined driving path (TR).

9. The method as claimed in one of the preceding claims, characterized in that the received sensor signals (SIG1(t)) exclusively comprise ultrasonic sensor signals.

10. The method as claimed in one of the preceding claims, characterized in that the number of different determination methods (V1, V2) comprises at least one second determination method in which, on the basis of the number of sensor signals (SIG1 (t )) a feature recognition is carried out and a digital map of the area is determined using recognized features.

11. The method according to any one of the preceding claims, characterized in that the method is carried out exclusively when the vehicle (100) has a speed of less than or equal to 15 km/h, preferably less than or equal to 10 km/h, preferably less than or equal to 7 km/h, more preferably less than or equal to 5 km/h.

12. A computer program product comprising instructions which, when the program is executed by a computer, cause the latter to execute the method according to any one of claims 1-11.

13. Driver assistance system (110) for a vehicle (100), with a receiving unit (112) for receiving a driving condition of the vehicle (100) indicative driving condition sensor signal (SIG0 (t)) at a number of different times (tO - 15), and for receiving a number of sensor signals (SIG1 (t)) indicative of an area (200) of the vehicle (100) at a number of different points in time (t0 - 15), a detection unit (114) for detecting a number of objects (210, 211) in the surroundings (200) of the vehicle as a function of a first number of sensor signals (SIG1 (t)) which were received at a first point in time, a determination unit ( 116) for determining a position (POS) and a motion vector (VEC) for a detected object (210) in dependence on the first number of sensor signals (SIG1 (t)) and a second number of sensor signals (SIG1 (t)), the was received at a second point in time following the first point in time, using a plurality of different determination methods (V1, V2), wherein different determination methods (V1, V2) of the plurality have a different computing effort, and an output unit (118) for outputting a warning signal , when a potential collision of the vehicle (100) with the detected object (210, 211) on the basis of the driving condition sensor received at a specific point in time (tO-15). signal (SIG0(t)) and the position (POS) determined for the detected object (210, 211) and its motion vector (VEC).

14. Vehicle (100) with a number of environment sensor units (120, 130) for detecting an environment (200) of the vehicle (100) and for outputting a respective sensor signal (SIG1(t) and with a driver assistance system (110) according to claim 13.

15. Vehicle according to claim 14, characterized in that the environmental sensor units (120, 130) exclusively comprise ultrasonic sensors.

16. Vehicle according to claim 14 or 15, characterized in that the vehicle (100) has a mass of more than 2.5 tons and/or a length of more than 5 meters.