DE102020204044A1 - Procedure for warning living beings - Google Patents

Abstract

In a method (V) for warning living beings (60) in an environment (2) of a vehicle (1), environment sensors (3) of the vehicle (1) monitor the environment (2) of the vehicle (1), whereby environment data (50) to be generated. Driving dynamics sensors (4) of the vehicle (1) monitor the driving dynamics of the vehicle (1), whereby driving dynamics data (52) are generated. The environment data (50) are evaluated and, based on the evaluated environment data (50 *), at least one living being (60) is detected in the environment (2) of the vehicle (1) and classified according to its species (51). A trajectory (53) of the at least one living being (60) is determined on the basis of the evaluated environment data (50 *). The driving dynamics data (52) are evaluated and, based on the evaluated driving dynamics data (52 *), a trajectory (54) of the vehicle (1) is determined. Starting from the trajectory (53) of the at least one living being (60) and starting from the trajectory (54) of the vehicle (1), a collision probability (55) between the vehicle (1) and the at least one living being (60) is determined. If the collision probability (55) reaches or exceeds a threshold value (56), based on the result of the classification, at least one signal (58) specific for the at least one living being (60) is output in order to warn the at least one living being (60), see above that the collision is avoided.

Description

The present invention relates to a method for warning living beings, an evaluation device, a computer program product and a vehicle.

The driver of a vehicle is responsible for the detection and reaction in potentially dangerous situations with e.g. B. Responsible for animals or people. Depending on the equipment of the vehicle, there are few options such. B. to recognize the situation correctly and in good time in the dark. The options for warning are also very limited and, in the worst case scenario, can worsen the situation. Damage caused by e.g. B. Wildlife accidents or unsuccessful evasive maneuvers lead to considerable personal and economic damage. Currently, drivers are driven by the infrastructure, e.g. B. warned by warning signs.

the end DE 10 2011 121 284 A1 a method for generating artificial driving noises of a motor vehicle is known. The artificial driving noises are generated by a sound generating unit and emitted acoustically to the outside, the properties of the driving noises being dependent on the detected objects in the vicinity of the vehicle.

the end DE 10 2011 102 759 A1 a method for influencing wild animals is known. A warning is sent to the wild animals.

the end DE 10 2011 102 760 A1 a vehicle with a safety device for avoiding the consequences of a collision with a wild animal is known.

Further systems and methods for warning animals and people in the vicinity of a vehicle are known from WO 2019 180 551 A1 , DE 10 2017 115 011 A1 , US 9481366 B1 , US 9481367 B1 , DE 10 2005 035 040 A1 , DE 10 2017 203 899 A1 , DE 10 2016 215 470 A1 , DE 10 2016 208 582 A1 , DE 10 2015 224 739 A1 , EP 2944516 A1 , DE 10 2014 005 464 A1 , DE 10 2009 027 911 A1 , EP 3404640 A1 , DE 10 2017 214 168 A1 , DE 10 2017 113 135A1 , DE 10 2016 117 700 A1 , DE 10 2014 110 958 A1 , KR 101 929 806 B1 .

Based on the prior art, the present invention is based on the object of proposing an improved method for warning living beings, that is to say animals and people, in the vicinity of a vehicle. The warning should be specifically tailored to the respective living being.

Based on the aforementioned object, the present invention proposes a method for warning living beings with the features of claim 1, an evaluation device with the features of claim 9, a computer program product with the features of claim 10 and a vehicle with the features of claim 11. Further advantageous refinements and developments emerge from the subclaims.

In a method for warning living beings in the surroundings of a vehicle, surroundings sensors of the vehicle monitor the surroundings of the vehicle, as a result of which surroundings data are generated. Driving dynamics sensors of the vehicle monitor a driving dynamics of the vehicle, whereby driving dynamics data are generated. The surroundings data are evaluated and, based on the evaluated surroundings data, at least one living being is detected in the surroundings of the vehicle and classified according to its species. A trajectory of the at least one living being is determined on the basis of the evaluated environment data. The driving dynamics data are evaluated and a trajectory of the vehicle is determined on the basis of the evaluated driving dynamics data. Starting from the trajectory of the at least one living being and starting from the trajectory of the vehicle, a collision probability between the vehicle and the at least one living being is determined. If the probability of a collision reaches or exceeds a threshold value, based on the result of the classification, at least one signal specific for the at least one living being is output in order to warn the at least one living being so that the collision is avoided.

The vehicle can be designed as a land vehicle. Preferably the vehicle is a land vehicle, e.g. B. a car, commercial vehicle, bus, people mover or the like. The vehicle is designed in such a way that it can perform automated functions. For example, automated functions of the vehicle can be autonomous or semi-autonomous driving. The vehicle can e.g. B. Carry out automated functions from level 4. Level 4 refers to the autonomy level according to SAE J3016.

A living being is defined here as an animal or as a person. Both the animal and the person are able to perceive their respective surroundings and to react to them. Animals that are to be warned by means of the method described here are, for example, haired game, game birds, large game, farm animals, dogs, cats, amphibians, crustaceans, etc. Insects, arachnids and worms are not covered by the method. Plants, fungi, bacteria, sponges, viruses are also excluded.

The surroundings of the vehicle are defined as the immediate surroundings of the vehicle in which it is moving. The vehicle travels along a route in this environment. For example, the vehicle can move autonomously along the route.

The vehicle has several environment sensors which are set up to monitor the environment of the vehicle. The surroundings sensors are arranged on the vehicle in such a way that they enable an all-round view as far as possible. Alternatively, monitoring can only take place in the area of the front of the vehicle. These environment sensors can be designed as imaging sensors, as acoustic sensors and / or as olfactory sensors. For example, the vehicle can have multiple imaging sensors, e.g. B. one or more cameras, one or more radar sensors, one or more lidar sensors. The camera can also be designed as an infrared camera, for example. For example, the vehicle may have one or more acoustic sensors, e.g. B. Microphones that can capture the area around the vehicle. For example, the vehicle may have one or more olfactory sensors, e.g. B. artificial noses that can detect the surroundings of the vehicle. Of course, the vehicle can have a combination of the aforementioned environment sensors.

The environment sensors are used to generate environment data that are based on the shape of the respective environment sensors. The environment data thus include, depending on the configuration of the environment sensors, optical environment data and / or acoustic environment data and / or olfactory environment data. These environment data are merged with one another by means of sensor data fusion, so that not only the environment data of a single environment sensor or a single environment sensor type are used to record the environment of the vehicle.

At the same time, the vehicle's driving dynamics sensors monitor the vehicle's driving dynamics, which generates driving dynamics data. For this purpose, the vehicle has the driving dynamics sensors. The driving dynamics sensors can be designed, for example, as speed sensors, as acceleration sensors, as steering angle sensors, as yaw rate sensors, as inclination sensors or as other driving dynamics sensors that are suitable for monitoring the driving dynamics of the vehicle. The driving dynamics describe a current speed and acceleration of the vehicle as well as a current direction of the vehicle while the vehicle is in motion. In other words, the driving dynamics can be used to determine in which direction the vehicle is moving with which speed and acceleration.

The surroundings data are evaluated and, based on the evaluated surroundings data, the at least one living being in the surroundings of the vehicle is detected and classified according to its species. The evaluation is carried out by means of an evaluation device which the vehicle has. The evaluation device is connected to the surroundings sensors and to the driving dynamics sensors in such a way that data and signals can be exchanged. This connection can be wireless or wired. For this purpose, the evaluation device has at least one interface, so that this connection is made possible. The evaluation device can be designed, for example, as an ECU or a domain ECU.

The evaluation of the environmental data takes place, for example, by means of an artificial intelligence that uses a trained artificial neural network that is trained to recognize living beings and to assign a species to these living beings. The trained artificial neural network is designed in such a way that it is further trained on the basis of experiences made by the vehicle during its operation. If, for example, a living being has been correctly identified and correctly assigned to a species, this information is used to further train the trained artificial neural network.

If a living being is recognized, it is assigned to a species. In other words, it is established which living being it is, for example a human being, a deer, a dog, a wild boar, etc. Species is to be understood in the biological sense, equivalent to the concept of species.

The trajectory of the at least one living being is then determined on the basis of the evaluated environment data. This is done by means of the evaluation device, which can use an algorithm or artificial intelligence for this purpose. The trajectory of the living being is determined on the basis of a recorded speed, a recorded acceleration, and a recorded direction of the living being. In other words, it is observed how the living being moves and, based on this, predicts where the living being will move. The assigned species of the living being also flows into the determination of the trajectory. For example, specific movement patterns for this species are included in the trajectory determination. These movement patterns can be called up, for example, from an external database, for example a cloud, and / or stored in a memory of the vehicle.

The driving dynamics data are evaluated and a trajectory of the vehicle is determined on the basis of the evaluated driving dynamics data. This evaluation of the driving dynamics data takes place by means of the evaluation device or by means of a further evaluation device. The trajectory of the vehicle shows in which direction and at what speed and acceleration the vehicle is moving and will be moving. In addition, data from route planning or a route profile can be incorporated into the determination of the vehicle trajectory. This data material can e.g. B. in a memory of the evaluation device of the vehicle and / or in an external data memory which the vehicle can call up.

Starting from the trajectory of the at least one living being and starting from the trajectory of the vehicle, a collision probability between the vehicle and the at least one living being is determined. In other words, the probability of a collision between the vehicle and the living being occurring is calculated if both maintain their trajectory. This calculation is also known as accident prediction. Accident prediction methods are known and available on the market.

If the collision probability reaches or exceeds the threshold value, based on the result of the classification, a specific signal is output for the at least one living being in order to warn the at least one living being so that the collision is avoided. The threshold value for the likelihood of collision is set at the factory, for example. The threshold value is chosen so that it is not too low, so that not every sighting of a living being leads to a reaction. In addition, the threshold value is selected in such a way that it is not too high, so that the risk of collision is kept low. For example, the threshold value can be set to a collision probability of 50%, 60%, 70% or higher. Of course, the threshold value can also be set lower or higher. In this context, threshold value does not mean that it is a global limit value that can never be exceeded.

Only when the threshold value has been reached or exceeded is the at least one signal that is specific to the at least one living being output to the surroundings of the vehicle. This signal serves as a warning so that the at least one living being is made aware of the impending collision and is thereby deterred so that the living being does not move any further on its predicted trajectory. Because of the signal, the living being evades, turns around or stops, for example. The trajectory of the living being thus changes and no longer corresponds to the predicted trajectory.

The output of the signal is controlled and initiated by means of the evaluation device. In order to be able to output the signal, the vehicle has at least one signal device. This is connected to the evaluation device so that data and signals can be exchanged. This connection can be wireless or wired. For this purpose, the evaluation device has at least one interface, so that this connection is made possible. The at least one signaling device can be designed, for example, as a vehicle system of the vehicle that is regularly present in the vehicle, e.g. B. a vehicle exterior lighting, a horn, a windshield washer or the like. Alternatively or in addition to this, the signaling device can be designed as an additional vehicle system, e.g. B. as a loudspeaker, as a lighting system, as a fragrance dispenser, as a vibration generator, etc.

The signal is specific to the living being and is selected on the basis of the result of the classification. For example, specific signals for each species can be stored in a memory of the vehicle and / or in an external database. Each signal can be in the form of an acoustic signal, an optical signal, an olfactory signal and / or a physical signal, for example. Of course, the signal can also be formed as a combination of the signal types just mentioned. For example, an acoustic signal can be combined with an olfactory signal. For example, a visual signal can be combined with an acoustic signal.

An optical or visual signal is, for example, a light signal, e.g. B. fading in or dimming the vehicle exterior lighting, targeted lighting of an area in the vicinity of the vehicle, etc. An acoustic signal is, for example, a warning tone, a tone sequence, a melody, a noise, an animal sound, an announcement, etc. The acoustic signal Signal can, for example, be in the ultrasound range, in the infrasound range or in the sound range that can be perceived by humans. An olfactory signal is, for example, an odor, for example an animal odor, or a fragrance that is emitted by the vehicle into its surroundings. A physical signal is, for example, a signal that is emitted to the surroundings of the vehicle by means of physical objects, e.g. B. a water jet, a pressure wave or a cloud of smoke. The water jet can, for. B. can be implemented via a windshield washer system of the vehicle, which can be set so that the The water jet does not hit the window, but the area around the vehicle.

For example, the sound of a predator, which is the natural enemy of the detected creature, can be output as an acoustic signal to warn certain animal species. B. howling wolves in deer, horses or wild boars, buzzard screams in rabbits, etc. Additionally or alternatively, a flash of light can be output as an optical signal reminiscent of flashes. Additionally or alternatively, a predator odor can be emitted as an olfactory signal.

As a result, every living being that is around the vehicle and with which there is a high probability of a collision can be deterred in a targeted manner.

The advantage of the method presented here is that it is possible in a simple manner to prevent collisions between the vehicle and living beings without changing or adapting the trajectory of the vehicle. The vehicle can continue its journey uninterrupted, as living beings are specifically warned and scared away before a collision can occur.

According to a further-developing embodiment, data on the surroundings of other vehicles that are moving on the same route as the vehicle are also included in the evaluation of the data on the surroundings of the vehicle. The other vehicles do not have to move in the same direction as the vehicle. All they have to do is drive the same route without a long interval. The other vehicles also have surroundings sensors that monitor the surroundings of these additional vehicles. If at least one further vehicle detects a living being in its environment while it is traveling on the same route as the vehicle, the further vehicle sends this information to the vehicle. This is done using C2C communication. Alternatively or in addition to this, the further vehicle can send this information to an external database, e.g. B. to a cloud. The vehicle can then call up information about the living being from this cloud. This is done using C2X communication.

This ensures that the vehicle is always informed about all living beings that are in its environment in order to react to them, even if the vehicle cannot or has not detected all living beings itself.

According to a further development, the species classification of the at least one living being also incorporates classification data from at least one external database in order to check the plausibility of the species classification of the at least one living being. These classification data are e.g. B. formed by means of optical classification data and / or acoustic classification data and / or olfactory classification data. The classification data of the external database can be continuously expanded, for example, by means of classification data from the vehicle and from other vehicles that communicate them to the external database. On the basis of these classification data from the external database, it is possible to decide more quickly which species it is when a living being is detected by the vehicle. The artificial intelligence of the vehicle no longer has to go through all the decision-making steps of the trained artificial neural network, but can therefore fall back on empirical values from the external database. This makes the classification of the species of the living being more accurate and faster.

According to a further developing embodiment, trajectory information that is specific for the species of the at least one living being from at least one external database and / or from other vehicles is also included in the determination of the trajectory of the at least one living being in order to determine the trajectory of the at least one living being to check plausibility. For example, empirical values from the other vehicles or the vehicle that were communicated to the external database can be stored in the external database. As an alternative or in addition to this, the further vehicles can communicate the additional trajectory information directly to the vehicle. This is advantageous when another vehicle has already determined the trajectory of the same living being that was captured by the vehicle and whose trajectory is still to be determined by the vehicle.

This additional trajectory information is related to the species of the living being. Such trajectory information contains movement patterns typical for the living being, e.g. B. jumping locomotion in red deer, lacing locomotion in foxes, hooking locomotion in rabbits or hares etc.

The advantage here is that this additional trajectory information makes the determined trajectory of the living being more precise. As a result, the probability of a collision can also be determined in a more precise manner.

According to a further developing embodiment, the environment data include optical environment data and / or acoustic environment data and / or olfactory environment data. This depends on the Formation of the vehicle's environment sensors. This has already been explained.

According to a further developing embodiment, the signal specific for the at least one living being is selected based on empirical values of the vehicle and based on empirical values of other vehicles and / or based on specific signals stored in the external database. The signal specific for the living being is thus a signal which has already been tested on at least one second living being of the same species and which has warned the living being in an effective manner and possibly frightened it so that there was no collision. For example, the signal may have been determined to be effective on an earlier journey by the vehicle and also be determined to be effective by a further vehicle.

The signal can in turn comprise a combination of several sensory impressions for the living being. For example, an optical signal, e.g. B. a flash of light are sent out together with an acoustic signal, namely the animal sound of a predator. For example, to warn farm animals, e.g. B. cows, an optical signal, e.g. B. a flash of light, emitted together with a physical signal, e.g. B. a jet of water directed at the cow.

In addition, the specific signal can be optimized and / or adapted by modifying and observing the reaction of the at least one living being. If the reaction or the movement of the at least one living being does not change or does not meet expectations, the specific signal can be adjusted, e.g. B. an intensity of the signal can be increased. Furthermore, statistical data on accidents involving wildlife on the route, which are stored in the external database, can be used to infer the effective specific signal. These statistical data can, for example, depend on a time, the position of the sun, a rutting season, a migration or migration time or the like.

It is advantageous here that only those specific signals are used by the vehicle for warning purposes that have already been classified as effective. As a result, the living being can be warned more quickly than if the vehicle alone selects the signal.

According to a further developing embodiment, the vehicle forwards the specific signal to the external database and / or to other vehicles if the specific signal was effective. The specific signal is effective when the vehicle has been able to warn and possibly deter the living being by means of the signal, so that a collision between the vehicle and the living being could be prevented. If the vehicle has used a signal that is effective for a species, this is communicated together with the species classification to the external database and stored there so that it can be called up by the vehicle itself and by other vehicles. Additionally or alternatively, the vehicle communicates the effective signal together with the species classification to the other vehicles. This makes it possible for the external database to grow steadily and to be filled with empirical values from the vehicle.

According to a further developing embodiment, the signal specific for the at least one living being is formed as at least one of the following specific signals or as a combination of the following specific signals: acoustic signal; optical signal; olfactory signal; physical signal. This has already been explained.

The evaluation device for the vehicle is set up to be connected to the environment sensors of the vehicle and to the driving dynamics sensors of the vehicle and to the signaling device of the vehicle, the evaluation device having means to carry out the method that has already been described in the previous description. This means that the evaluation device is connected to the just mentioned system and devices of the vehicle when the evaluation device is used in a vehicle. The evaluation device, the driving dynamics sensors, the environment sensors, and the signaling device have already been described in the previous description. Likewise, the connection of the evaluation device with the just mentioned devices of the vehicle has already been described in the previous description.

The evaluation device has means to carry out the method that has already been described in the previous description. These means can be implemented, for example, as a computer program product that runs on the evaluation device. The procedure has already been described.

The computer program product comprises commands which, when the program is executed by the evaluation device already described, execute the method which has also already been described. The computer program product can comprise a program code that contains these commands. The program code can be embodied, for example, on a data carrier or as a downloadable data stream.

The vehicle has the evaluation device that has already been described. In addition, the Vehicle on the environment sensors, driving dynamics sensors and the signal device, which have already been described. The evaluation device is connected to the environment sensors and the driving dynamics sensors and the signal device. The vehicle is set up to perform automated functions, e.g. B. to drive autonomously. This has already been described.

• 1 eine schematische Darstellung eines Fahrzeugs nach einem Ausführungsbeispiel,
• 2 eine schematische Darstellung eines Verfahrens, das von dem Fahrzeug aus 1 durchgeführt wird.

Various exemplary embodiments and details of the invention are described in more detail with the aid of the figures explained below. Show it:

• 1 a schematic representation of a vehicle according to an embodiment,
• 2 a schematic representation of a method from the vehicle 1 is carried out.

1 shows a schematic representation of a vehicle 1 according to an embodiment. The vehicle 1 moves in its environment 2 . The vehicle 1 has: an evaluation device 10 , a signaling device 11 , several environmental sensors 3 , of which only one environment sensor for a better overview 3 is shown, and several driving dynamics sensors 4th , of which only one driving dynamics sensor for a better overview 4th is shown. The evaluation device 10 is with the signaling device 11 , with the environment sensors 3 and with the driving dynamics sensors 4th connected so that data and signals can be exchanged. The respective connections can be wireless or wired.

The environment sensors 3 serve the environment 2 of the vehicle 1 capture. The environment sensors 3 generate environmental data during this acquisition. The driving dynamics sensors 4th serve to improve the driving dynamics of the vehicle 1 to monitor. The driving dynamics sensors 4th generate vehicle dynamics data during this monitoring. Both the environment data and the driving dynamics data are sent to the evaluation device 10 forwarded, which evaluates this.

On the basis of the evaluated environment data it can be determined that in the environment 2 of the vehicle 1 a living being 60 , here a deer, is located. The evaluation device 10 uses an artificial intelligence, which in turn makes use of a trained artificial neural network, in order to use the environmental data as a starting point for the living being 60 to be assigned to a species, i.e. the living being 60 to be classified according to its species. This is done in 2 explained in more detail.

To the living being 60 in front of the vehicle 1 Warn is from the vehicle 1 by means of its signaling device 11 a specific signal 58 output that is matched to the species of the living being 60 . The procedure for this is described in 2 explained in more detail.

The vehicle 1 also has a communication device, not shown here, which is set up to work with external systems, for example an external database D. and / or other vehicles that are in the vicinity of the vehicle 1 located to communicate. This communication takes place on the way of the C2X or. C2C communication. The external database D. can be designed as a cloud, for example.

The vehicle 1 can from the external database D. z. B. Classification data on the species of the living being 60 and / or information on trajectories specific to the species is obtained. In addition, the vehicle can 1 from the external database D. specific signals 58 retrieve that for the warning of the species of the classified living thing 60 are effective. Furthermore, the vehicle can 1 its ascertained classification data and trajectory data with regard to the living being 60 to the external database D. forward onto. In addition, the vehicle can 1 to the external database D. forward what specific signal 58 was used to the living being 60 to warn and whether this specific signal 58 was effective.

2 shows a schematic representation of a method V from the vehicle 1 is carried out. The procedure V runs continuously and repeatedly while the vehicle is in motion. That is, the procedure V After it has been used once to warn a living being, it can be run through again to warn another or the same living being in the same or in a different way.

In a first step 101 The vehicle's environment sensors monitor the vehicle’s environment, creating environment data 50 to be generated. This environmental data 50 are based on the shape of the respective environment sensors. For example, optical and / or acoustic and / or olfactory environment data can be used 50 be. The environment data 50 the individual environmental sensors are merged with one another by means of sensor data fusion.

In a second step 102 who took the first step 101 then the environment data 50 evaluated. This results in evaluated environmental data 50 * obtain. Based on the evaluated environmental data 50 * the living being in the vicinity of the vehicle is detected and according to its species 51 classified. This evaluation, as well as the classification, is carried out by means of the evaluation unit of the vehicle, which for this purpose uses a computer program product that uses artificial intelligence may have. Optionally, additional environmental data 70 of other vehicles that move in the same environment as the vehicle are included in the evaluation. For example, another vehicle can provide additional visual environment data 70 deliver that with the surrounding data 50 of the vehicle are merged. In this way, for example, hidden living beings can be better detected.

Again, additional classification data 71 flow into the species classification of the at least one living being from the external database in order to check the plausibility of the species classification of the at least one living being. This additional classification data 71 can for example for the species 51 include typical features that are used in the evaluation of the environmental data 50 can be used. For example, the additional classification data 71 contain optical classification data and / or acoustic classification data and / or olfactory classification data which are specific for the species 51 . This allows the classification of the species 51 done in a faster way.

In a third step 103 who took the second step 102 is based on the evaluated environmental data 50 * a trajectory 53 of the at least one living being determined. This trajectory 53 describes where the living being will move at what speed and with what acceleration. In other words, a behavior of the living being is predicted. In determining the trajectory 53 of the living being can optionally have additional trajectory information 72 that is for the species 51 of the at least one living being are specific to flow from the external database and / or from other vehicles. This allows the trajectory 53 of the living being can be determined in a faster way, because it can be based on empirical values with the classified species 51 can be used. For example, specific movement patterns can be stored in the external database for each species, which can be called up by the vehicle or can be communicated to the vehicle.

In a fourth step 104 that runs parallel to the first step 101 can run, the driving dynamics sensors of the vehicle monitor the driving dynamics of the vehicle, creating driving dynamics data 52 to be generated. The driving dynamics data 52 depend on the type of the respective driving dynamics sensors. The driving dynamics data 52 can therefore include speed data, acceleration data, steering angle data, roll data, yaw rate data or the like.

In a fifth step 105 who passed the fourth step 104 connects, the driving dynamics data 52 evaluated, whereby evaluated driving dynamics data 52 * be won. This evaluation, as well as the classification, takes place by means of the evaluation unit of the vehicle, which uses a computer program product for this purpose. Based on the evaluated driving dynamics data 52 * becomes a trajectory 54 of the vehicle. In other words, it is determined where the vehicle will move at what speed and with what acceleration. In determining the trajectory 54 of the vehicle

In a sixth step 106 who at the third step 103 and to the fifth step 105 is based on the trajectory 53 of the classified living being and based on the trajectory 54 of the vehicle a collision probability 55 determined between the vehicle and the living being. This determination of the likelihood of collision 55 takes place by means of the evaluation unit, which uses algorithms for accident prediction that are already available on the market.

In a seventh step 107 the one who took the sixth step 106 then becomes the probability of a collision 55 with a threshold 56 compared, which was set for example at the factory. The threshold 56 sets the value of the collision probability 55 determines from which the collision probability 55 is so high that the vehicle must take action to prevent the collision. If the threshold 56 is reached or even exceeded, is then based on the result of the classification in step 102 at least one signal specific to the living being 58 issued to warn the at least one living being so that the collision is avoided. The specific signal 58 in other words, depends on the species 51 of the living being. The specific signal can be formed, for example, as an optical signal and / or an acoustic signal and / or a physical signal and / or an olfactory signal or as a combination thereof. The specific signal 58 is chosen in such a way that the living being is warned and at best frightened, so that it is diverted from its predicted trajectory 53 deviates. This can effectively prevent the collision between the vehicle and the living being. The vehicle does not have to start or stop an evasive maneuver, but can continue driving without deviating from its trajectory. Thus, the driving operation is not disturbed.

Optionally, the vehicle can if it has the specific signal 58 effectively used the specific signal to warn the living being 58 coupled with information about the species 51 of the living being to the external database or to others Forward vehicles. This means that empirical values can be shared.

List of reference symbols

1

vehicle

2

surroundings

3rd

Environment sensors

4th

Driving dynamics sensors

10

Evaluation device

11

Signaling device

50

Environment data

50 *

evaluated environmental data

51

species

52

Driving dynamics data

52 *

evaluated driving dynamics data

53

Trajectory of the living being

54

Vehicle trajectory

55

Probability of collision

56

Threshold

58

specific signal

60

Creature

70

additional environment data

71

additional classification data

72

additional trajectory information

101

first step

102

second step

103

Third step

104

fourth step

105

fifth step

106

sixth step

107

seventh step

D.

external database

V

procedure

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• DE 102011121284 A1 [0003]
• DE 102011102759 A1 [0004]
• DE 102011102760 A1 [0005]
• WO 2019180551 A1 [0006]
• DE 102017115011 A1 [0006]
• US 9481366 B1 [0006]
• US 9481367 B1 [0006]
• DE 102005035040 A1 [0006]
• DE 102017203899 A1 [0006]
• DE 102016215470 A1 [0006]
• DE 102016208582 A1 [0006]
• DE 102015224739 A1 [0006]
• EP 2944516 A1 [0006]
• DE 102014005464 A1 [0006]
• DE 102009027911 A1 [0006]
• EP 3404640 A1 [0006]
• DE 102017214168 A1 [0006]
• DE 102017113135 A1 [0006]
• DE 102016117700 A1 [0006]
• DE 102014110958 A1 [0006]
• KR 101929806 B1 [0006]

Claims (11)

Method (V) for warning living beings (60) in an environment (2) of a vehicle (1), wherein - Environment sensors (3) of the vehicle (1) monitor the environment (2) of the vehicle (1), whereby environment data (50) are generated, - Driving dynamics sensors (4) of the vehicle (1) monitor the driving dynamics of the vehicle (1), whereby driving dynamics data (52) are generated, - the environment data (50) are evaluated and, based on the evaluated environment data (50 *), at least one living being (60) is detected in the environment (2) of the vehicle (1) and classified according to its species (51), - On the basis of the evaluated environment data (50 *), a trajectory (53) of the at least one living being (60) is determined, - The driving dynamics data (52) are evaluated and, based on the evaluated driving dynamics data (52 *), a trajectory (54) of the vehicle (1) is determined, - starting from the trajectory (53) of the at least one living being (60) and starting from the trajectory (54) of the vehicle (1), a collision probability (55) between the vehicle (1) and the at least one living being (60) is determined, - if the collision probability (55) reaches or exceeds a threshold value (56), based on the result of the classification, at least one signal (58) specific for the at least one living being (60) is output in order to warn the at least one living being (60), so that the collision is avoided. Procedure (V) according to Claim 1 wherein additional environment data (70) of other vehicles moving on the same route as the vehicle (1) are included in the evaluation of the environment data (50) of the vehicle (1). Method (V) according to one of the preceding claims, wherein additional classification data (71) of at least one external database (D) flow into the species classification of the at least one living being (60) in order to assign the species classification of the at least one living being (60) check plausibility. Method (V) according to one of the preceding claims, wherein in the determination of the trajectory (53) of the at least one living being (60) additional trajectory information (72) which is specific for the species (51) of the at least one living being (60) , flow in from at least one external database (D) and / or from further vehicles in order to check the plausibility of the trajectory (53) of the at least one living being (60). Method (V) according to one of the preceding claims, wherein the environment data (50) comprise optical environment data (50) and / or acoustic environment data (50) and / or olfactory environment data (50). Method (V) according to one of the preceding claims, wherein the signal (58) specific for the at least one living being (60) is based on empirical values of the vehicle (1) and based on empirical values of other vehicles and / or based on in the external database (D ) stored specific signals (58) is selected. Method (V) according to one of the preceding claims, wherein the vehicle (1) forwards the specific signal (58) to the external database (D) and / or to other vehicles if the specific signal (58) was effective. Method (V) according to one of the preceding claims, wherein the signal (58) specific for the at least one living being (60) is formed as at least one of the following specific signals (58) or as a combination of the following specific signals (58): acoustic Signal; optical signal; olfactory signal; physical signal. Evaluation device (10) for a vehicle (1), wherein the evaluation device (10) is set up with the environment sensors (3) of the vehicle (1) and with the driving dynamics sensors (4) of the vehicle (1) and with a signaling device (11) ) of the vehicle (1) to be connected, the evaluation device (4) having means to carry out the method (V) according to one of the preceding claims. Computer program product comprising commands which, when the program is executed by an evaluation device (10) Claim 9 , the method (V) according to one of the Claims 1 until 8th carry out. Vehicle (1) having an evaluation device (10) according to Claim 9 , Environment sensors (3), driving dynamics sensors (4) and a signaling device (11), the evaluation device (10) being connected to the environment sensors (3) and the driving dynamics sensors (4) and the signaling device (11), the vehicle (1) is set up to perform automated functions.

Images