DE102018200381A1 - Method for situation-dependent adaptation of a motor vehicle and motor vehicle - Google Patents

Abstract

The invention relates to a method for situation-dependent adaptation of a motor vehicle (1) with the following steps:
a) detecting sensor data by means of at least one sensor (4, 5, 6, 7, 8, 9, 10, 11) of the motor vehicle (1);
b) evaluating the sensor data and providing the result of the evaluation;
c) generating a control signal depending on the result of the evaluation; and
d) adjusting the motor vehicle (1) based on the generated control signal.

Description

The invention relates to a method for situation-dependent adaptation of a motor vehicle. Sensor data are detected by means of at least one sensor of the motor vehicle. Furthermore, the sensor data are evaluated and a result of the evaluation is provided. The invention also relates to a motor vehicle which is designed to carry out a corresponding method.

Vehicles are currently usually driven manually and assistance systems make recommendations to a driver of the vehicle. Upon activation of the assistance systems by the driver, these can then intervene on the drive train of the vehicle to reduce consumption.

An analysis of the driver is currently at security features such. As in possibility recognition systems to warn the driver inattention or to recommend the driver a break.

The disadvantage is that the vehicle or the drive train of the vehicle is currently not sufficiently adapted to a behavior of the driver. This leads to increased fuel consumption in combustion vehicles and a higher battery load in electric vehicles.

It is an object of the invention to provide a method and a motor vehicle in which or with which a motor vehicle can be operated more energy-efficient.

The object is achieved by a method and a motor vehicle according to the independent patent claims.

1. a) Erfassen von Sensordaten mittels zumindest einem Sensor des Kraftfahrzeugs;
2. b) Auswerten der Sensordaten und Bereitstellen eines Ergebnisses der Auswertung;
3. c) Erzeugen eines Steuersignals abhängig von dem Ergebnis der Auswertung; und
4. d) Anpassen des Kraftfahrzeugs anhand des erzeugten Steuersignals.

In a method according to the invention, a motor vehicle is adapted to the situation. The following steps are carried out:

1. a) detecting sensor data by means of at least one sensor of the motor vehicle;
2. b) evaluating the sensor data and providing a result of the evaluation;
3. c) generating a control signal depending on the result of the evaluation; and
4. d) adjusting the motor vehicle based on the generated control signal.

The invention is based on the knowledge that the motor vehicle can be adapted to the situation and thus operated more energy-efficient.

By adjusting, in particular, a component of the motor vehicle, in particular a drive train of the motor vehicle, is adapted in such a way that after the adaptation, less energy is needed than before for the operation of the motor vehicle.

The drive train of the motor vehicle is adapted to the driver in particular to the respective driver in order to avoid unnecessary biasing of the drive train.

In particular, driver-specific patterns are identified or personalized, in particular to derive measures for optimizing the drive train of the motor vehicle.

For this purpose, the data or sensor data from the vehicle sensor cluster or the sensor and the vehicle network are analyzed or processed (inside and outside, eg inside: indoor cameras, microphone, built-in sensors, data from bus systems, eg mobile phone use or for Music consumption, outside: radar or lidar or camera, but also built-in sensors). By means of at least one algorithm, the previously preprocessed sensor data are analyzed in a first step and driver-specific patterns are recognized. This phase is also called learning phase. In order subsequently to adapt the drive train to the driver of the motor vehicle on the basis of the learned patterns, without producing disadvantages for the respective driver.

The algorithms used are preferably machine learning algorithms, such as a clustering method or a neural network. The machine learning algorithms learn driver-specific patterns and assign suitable energy-saving measures to these patterns. Furthermore, the recognized patterns are specifically assigned to the driver and stored. The detected patterns can also be made available to other drivers with similar behavior. So it can be formed a class of similar drivers.

It is preferably provided that in step b) a communicating driver is recognized in the motor vehicle and in step b) the motor vehicle is moved from a normal operating mode into a power saving operating mode. It may thus be, for example, that the sensor detects that the driver is on the phone, for example by actively using a hands-free device of the motor vehicle and / or a microphone of the motor vehicle and / or a camera of the motor vehicle. But it may also be that the communicating driver is recognized because the driver is in intensive conversation with his passenger, which, for example, from a camera of the motor vehicle is detected. As a result of the driver communicating, it can be assumed that the driver will not currently retrieve the full engine power of the motor vehicle, especially as long as the driver's communication lasts. The motor vehicle can thus be operated in the energy-saving operating mode, without this impairing the driving experience of the driver or disturbing the driver. In the energy-saving operating mode, the motor vehicle consumes less energy, ie, for example, less fuel or less electrical energy than is the case in the normal operating mode. For example, in the energy-saving operating mode, a higher gear may be engaged, or else, for example, cylinders may be switched off, or a lower voltage for the traction unit may be provided by a traction battery of the motor vehicle.

Furthermore, it is preferably provided that in step b) at least one passenger is detected and the motor vehicle is moved starting from a normal operating mode in a safety operating mode in which at least one safety-related function of the motor vehicle is triggered earlier by a driver assistance device of the motor vehicle than in the normal operating mode. As the passenger, for example, a passenger or a driver can be recognized on a back seat of the motor vehicle. In the safety operation mode, for example, a distance warning to a preceding vehicle is issued earlier than in the normal operation mode. For example, the distance warning can be output at a greater distance to the vehicle in front, as is the case in the normal operating mode. Also, for example, an emergency brake assist earlier, d. H. intervene at a greater distance to the vehicle ahead than would be the case in the normal operating mode. Otherwise, it is assumed that the driver wishes to operate the motor vehicle more safely or wants to operate the motor vehicle more cautiously when he is not alone, but the passenger is on board the motor vehicle.

In particular, it is envisaged that a child, in particular with an age of less than 14, will be recognized as a passenger. Especially with children, especially in small children, especially babies, the driver shows a heightened need to carry the passengers safer in the car. It is then detected by the sensor, for example by a camera and evaluation of the sensor data or on the images of the camera, that the child is as a passenger in the vehicle and the motor vehicle is moved from the normal operating mode in the safety mode and preferably while the passenger on board is operated in safety mode.

Further preferably, it is provided that in step b) a trailer is recognized on the motor vehicle and in step d) the motor vehicle is moved from a normal operating mode in a trailer operating mode, which at a speed of the motor vehicle between 0 km / h and 80 km / h more Traction performance than in normal operation mode provides. The trailer on the motor vehicle can be recognized for example by a trailer coupling sensor on a trailer hitch of the motor vehicle. The trailer can also be detected, for example, by means of a camera of the motor vehicle. The motor vehicle is then preferably operated in the trailer operating mode as long as the trailer is detected as being attached to the motor vehicle. In the trailer operating mode, the motor vehicle is preferably operated such that at lower speeds, in particular between 0 km / h and 80 km / h, more traction power is available than is the case in the normal operating mode. In the normal operating mode, for example, the motor vehicle has more traction power available than in trailer operating mode at more than 80 km / h. This is advantageous since a motor vehicle with a trailer usually does not travel faster than 80 km / h and therefore does not require any engine power even at a speed of more than 80 km / h. The motor vehicle therefore does not have to provide the traction power for more than 80 km / h, but can provide it for less than 80 km / h and, for example, be operated more energy-efficient or operated more safely, since the motor vehicle with better traction power is now better adapted to driving situations How to perform an overtaking maneuver on a highway more safely.

Furthermore, it is preferably provided that in step b) a severe weather condition is detected and in step d) the motor vehicle is moved from a normal operating mode into a severe operating mode, which at a speed of the motor vehicle between 0 km / h and 60 km / h more traction power than in normal operation mode. The storm can be present, for example, in heavy rainfall or hail. The severe weather can also be detected in snowfall or snowy roads. The increase in traction power in severe weather mode, for example, be charged, so that at more than 60 km / h less traction power is available than is the case in the normal operating mode. This is nevertheless clearly advantageous, since the motor vehicle is not particularly fast, especially not faster than 60 km / h is moved in detected severe weather, for example, the visibility for the driver and / or sensors of the motor vehicle, such as a radar sensor and / or is limited to a Lidarsensor, compared to the normal weather condition. Advantageous is the severe weather operating mode, since the motor vehicle is safely in contact with the road surface even in the case of severe weather and can be moved on difficult terrain.

Furthermore, it is preferably provided that in step b) an output of a driver assistance system of the motor vehicle revealing reaction of a driver of the motor vehicle is detected and in step d) an error check of the driver assistance system is performed. Thus, for example, it is provided that an analysis of the driver's reactions to the driver assistance system is detected by, for example, an interior camera or an interior microphone, which, for example, suggests that the driver is dissatisfied. As the denigrating reaction can be interpreted, for example, loud scolding or hitting against an interior trim of the motor vehicle. If the slanderous reaction of the driver is recognized, an error check of the driver assistance system is preferably carried out in step d). It is checked whether the driver assistance system is operated incorrectly or operated properly. In the event that an error is found, a sensitivity of the driver assistance system can be adjusted so that the driver assistance system intervenes less strongly in the operation of the motor vehicle or any hints to the driver outputs as was previously the case. Also, for example, the last action of the driver assistance system can be recorded while the denigrating driver's reaction is detected. The denigrating reaction together with the last action of the driver assistance system can then be transmitted to an evaluation device, for example outside the vehicle, to more accurately examine the operation of the driver assistance system and, if necessary, to adapt the driver assistance system of the motor vehicle and / or other motor vehicles to a driver assistance system of the same type as the driver assistance system ,

Furthermore, it is preferably provided that in step b) a start-stop automatic of the motor vehicle disparaging reaction of the driver is detected and in step d) a configuration of the start-stop system is adjusted. So it may be, for example, that the driver gives the denigrating reaction when the start-stop automatic of the motor vehicle off an internal combustion engine of the motor vehicle just because of a short stoppage of the motor vehicle. However, the standstill is so short, for example, less than two seconds, so that turning off the engine is not worth it, because after a short time already continued. If it is detected in step b) that the driver gets excited about the driver assistance system, in particular the start-stop system, by emitting the denigrating reaction, the configuration of the start-stop system is adapted. The adaptation of the configuration of the automatic start-stop system can take place, for example, by stopping the internal combustion engine only after one to two or three seconds of standstill. However, it may also be, for example, that the switching off of the internal combustion engine takes place only after a traffic event has been detected and, for example, it has been established that there are sufficiently many clues that justify switching off the internal combustion engine, since it is assumed that the internal combustion engine is at least three, for example Seconds remains off. This can be the case, for example, if it is detected that a traffic light has just switched to red.

Furthermore, it is preferably provided that the result of the evaluation of step b) is assigned to a driver class of a plurality of driver classes and the result of the evaluation is stored together with the assignment. By assigning the result of the evaluation to the driver class, another driver, who is similar to the driver with regard to certain properties, such as safety thinking and consumer awareness, can profit from the result of the evaluation in the future without the evaluation having to be carried out again. The other driver can then use, for example, only the result of the evaluation to operate his motor vehicle.

Furthermore, it is preferably provided that when evaluating the sensor data in step b) a driver-specific pattern with a classifier of machine learning is detected and the control signal is generated in step c) depending on the recognized pattern. The classifier of machine learning can be generated, for example, with a cluster method or a neural network. By evaluating the sensor data with the classifier of machine learning, the evaluation of the sensor data can be faster and more accurate. The classifier is preferably trained with training data from a plurality of reference drivers. Each known pattern, which is recognized by the classifier, is preferably associated with a respective control signal, which is generated in step c). Thus, the pattern can finally be determined by the classifier and the control signal can then be generated or selected.

The invention also relates to a motor vehicle. The motor vehicle according to the invention has at least one sensor and an evaluation unit. The motor vehicle is designed to carry out a method according to the invention.

The sensor is preferably designed as a radar sensor, camera, laser scanner, lidar sensor (lidar light detection and ranging), ultrasonic sensor, touch sensor and / or electronic nose.

Advantageous embodiments of the method according to the invention are to be regarded as advantageous embodiments of the motor vehicle. The subject components of the motor vehicle are each designed to carry out the respective method steps.

Further features of the invention will become apparent from the claims, the figures and the description of the figures.

Embodiments of the invention will be explained in more detail with reference to schematic drawings.

• 1 zeigt eine schematische Seitenansicht eines Ausführungsbeispiels eines erfindungsgemäßen Kraftfahrzeugs; und
• 2 eine schematische Darstellung eines Ablaufdiagramms zur Erzeugung eines Klassifikators des maschinellen Lernens und Detektieren eines fahrerspezifischen Musters mit dem Klassifikator.

Show it:

• 1 shows a schematic side view of an embodiment of a motor vehicle according to the invention; and
• 2 a schematic representation of a flowchart for generating a classifier of the machine learning and detecting a driver-specific pattern with the classifier.

In the figures, identical or functionally identical elements are provided with the same reference numerals.

1 shows an embodiment of a motor vehicle 1 with an electric motor 2 and an internal combustion engine 3 , Both the electric motor 2 as well as the internal combustion engine 3 are designed as traction motors. The car 1 but only with the electric motor 2 alone or only with the combustion engine 3 be trained alone. The car 1 also has a radar sensor 4 , a lidar sensor 5 , a front outdoor camera 6 , a rear outdoor camera 7 , an interior camera 8th , an interior microphone 9 , a data bus system 10 , a trailer coupling sensor 11 a trailer hitch 12 of the motor vehicle 1 on.

At the trailer hitch 12 is a trailer according to the embodiment 13 attached. The car 1 However, you can do it without the trailer 13 be educated.

In the motor vehicle 1 There are according to the embodiment, a driver 14 and a passenger 15 , The driver 14 sits on a driver's seat 16 and the rider 15 sits according to the embodiment on a back seat 17 , The rider 15 but also on a passenger seat, not shown, of the motor vehicle 1 sit.

Whether the driver 14 now communicates, ie a conversation with the passenger 15 or via a telephone with an external person leads, for example, by the interior camera 8th and / or the interior microphone 9 and / or by, for example, a hands-free device 18 of the motor vehicle 1 data-transferring connected data bus system 10 be recognized.

The car 1 Then, according to the embodiment, upon detection of the communication of the driver 14 from a normal mode of operation 19 in a power save mode 20 added. In energy saving mode 20 becomes the internal combustion engine 3 and / or the electric motor 2 with a lower energy consumption than in the normal operating mode 19 operated. In energy saving mode 20 For example, a higher gear of a transmission 21 of the motor vehicle 1 be used. In energy saving mode 20 but also for example a traction battery 22 of the motor vehicle 1 with a lower output voltage than in normal operation mode 19 be configured.

The car 1 furthermore, according to the exemplary embodiment, has an emergency brake device designed as a brake system 23 on. The emergency brake device 23 will be in a safety mode of operation 24 of the motor vehicle 1 triggered earlier than in normal operating mode 19 , The car 1 becomes according to the embodiment of the normal operation mode 19 in the safety mode 24 offset if the passenger 15 in the motor vehicle 1 is recognized. The rider 15 For example, with the interior camera 8th and / or the interior microphone 9 and / or a pressure sensor in the rear seat 17 be recognized. According to the embodiment, the passenger is 15 younger than fourteen and thus a child.

Will the trailer 13 on the motor vehicle 1 for example, through the rear exterior camera 7 and / or the trailer coupling sensor 11 recognized, so will the motor vehicle 1 according to the embodiment of the normal operation mode 19 in a trailer operating mode 25 added. In trailer mode 25 is the motor vehicle 1 between 0 km / h and 80 km / h more traction power than in normal operating mode 19 to disposal. The more traction power is achieved, for example, by the trailer operating mode 25 less traction power is available at more than 80 km / h. However, it can also be provided, for example, that in trailer operating mode 25 an output voltage of the traction battery 22 is increased. Also, for example, in the transmission 21 be switched to a lower gear to one get better gear ratio between 0 km / h and 80 km / h.

According to a further embodiment, it may also be that a severe weather condition 26 in an environment 27 of the motor vehicle 1 is recognized. The storm condition 26 For example, by the front outdoor camera 6 be recognized. Will the storm condition 26 recognized, so will the motor vehicle 1 from the normal operation mode 19 in a storm mode of operation 28 added. In the storm mode 28 is the motor vehicle 1 then comparable to trailer mode 25 more traction power between 0 km / h and 60 km / h than in normal operating mode 19 to disposal.

In a further embodiment, it is provided that a disparaging reaction 29 the driver 14 for example, with the interior microphone 9 and / or the interior camera 8th is recognized. With the disparaging reaction 29 becomes the driver assistance system of the motor vehicle 1 denigrated. The driver assistance system can be used, for example, as a start-stop system 30 be educated. Will the denigrating reaction 29 the start-stop automatic 30 detected, so can a configuration 31 the start-stop automatic 30 be adjusted. So can the start-stop automatic 30 For example, be adapted so that the internal combustion engine 3 of the motor vehicle 1 is turned off less often or only for a longer time than this in the normal operating mode of the start-stop system 30 the case is or before adjusting the configuration 31 the case is.

2 shows a flowchart of a method for situation-dependent adjustment of the motor vehicle 1 , In one step S1 becomes the driver 14 of the motor vehicle 1 recognized or the driver 14 is assigned to an identification number. After the step S1 follows either a learning phase with a step S2 or a detection phase with one step S3 ,

Both in the step S2 as well as in the step S3 be sensor data of the motor vehicle 1 For example, from the interior camera 8th , the interior microphone 9 , the radar sensor 4 , the lidar sensor 5 , the front outdoor camera 6 , the outside rear camera 7 or vehicle data or backend data used by an off-board server.

In step S2 For example, the sensor data is processed by a machine learning method. Ie. the sensor data is used as training data and a driver profile in one step S4 assigned. In step S4 thus becomes in particular a classifier 32 generated.

The classifier 32 is then in the step S3 used a driver-specific pattern 33 to recognize.

In one step S5 a decision is made as to whether a pattern is recognized or whether no pattern is recognized. If no pattern is detected, follow the step S5 a step S6 , With the step S6 no further action follows. Will in step S5 the driver specific pattern 33 recognized, then follows a step S7 and / or a step S8 and / or a step S9 ,

In step S7 becomes a drive train of the motor vehicle 1 , in particular the electric motor 2 and / or the internal combustion engine 3 adjusted so that they are operated more energy efficient and / or in certain speed ranges more powerful.

In step S8 becomes a health state management (SOH - state of health) of the traction battery 22 customized.

In step S9 becomes a driver assistance system of the motor vehicle 1 , in particular the start-stop system 30 of the motor vehicle 1 customized.

In step S8 For example, it may be a predictive measure to relieve the traction battery 22 which lead to a longer life of the traction battery 22 leads.

Overall, by the method, a consumption saving and / or a longer life of the components of the motor vehicle 1 be achieved by driver-specific adaptation. But it can also be a higher acceptance of driver assistance systems of the motor vehicle 1 be reached and / or more comfort for the user or the driver 14 of the motor vehicle 1 be achieved. Furthermore, by adapting the driver assistance system of the motor vehicle 1 the safety of the motor vehicle 1 increase.

LIST OF REFERENCE NUMBERS

1

motor vehicle

2

electric motor

3

internal combustion engine

4

radar sensor

5

lidar

6

front outdoor camera

7

rear exterior camera

8th

Interior camera

9

Interior microphone

10

data bus system

11

Trailer coupling sensor

12

trailer hitch

13

pendant

14

driver

15

passengers

16

driver's seat

17

backseat

18

speakerphone

19

Normal operating mode

20

Energy-saving mode

21

transmission

22

traction battery

23

Emergency brake

24

Security mode

25

Trailer towing mode

26

bad weather condition

27

environment

28

Storm mode

29

disparaging reaction

30

Start-Stop technology

31

configuration

32

classifier

33

driver-specific pattern

Claims (11)

Method for situation-dependent adaptation of a motor vehicle (1) with the following steps: a) detecting sensor data by means of at least one sensor (4, 5, 6, 7, 8, 9, 10, 11) of the motor vehicle (1); b) evaluating the sensor data and providing a result of the evaluation; c) generating a control signal depending on the result of the evaluation; and d) adjusting the motor vehicle (1) based on the generated control signal. Method according to Claim 1 , characterized in that in step b) a communicating driver (14) in the motor vehicle (1) is detected and in step d) the motor vehicle (1), starting from a normal operating mode (19) is placed in a power saving mode (20). Method according to Claim 1 , characterized in that in step b) at least one passenger (15) is detected and the motor vehicle (1), starting from a normal operating mode (19) in a safety mode (24) is added, in which at least one safety-related function of the motor vehicle (1) is triggered by a driver assistance device of the motor vehicle earlier than in the normal operating mode (19). Method according to Claim 3 , characterized in that as the rider (15) a child is detected. Method according to Claim 1 , characterized in that in step b) a trailer (13) on the motor vehicle (1) is detected and in step d) the motor vehicle (1) is moved from a normal operating mode (19) to a trailer operating mode (25), which at one speed of the motor vehicle (1) provides more traction power between 0 km / h and 80 km / h than the normal operating mode (19). Method according to Claim 1 , characterized in that in step b) a severe weather condition (26) is detected and in step d) the motor vehicle (1) is shifted from a normal operating mode (19) into a severe operating mode (28), which at a speed of the motor vehicle (1) between 0 km / h and 60 km / h provides more traction power than the normal operating mode (19). Method according to Claim 1 , characterized in that in step b) an output of a driver assistance system of the motor vehicle (1) disparaging reaction (29) of a driver (14) of the motor vehicle (1) is detected and in step d) an error check of the driver assistance system is performed. Method according to Claim 1 or 7 , characterized in that in step b) a start-stop automatic (30) of the motor vehicle (1) denigrating reaction of a driver (14) is detected and in step d) a configuration (31) of the automatic start-stop ( 30) is adjusted. Method according to one of the preceding claims, characterized in that the result of the evaluation of step b) is assigned to a driver class of a plurality of driver classes and the result of the evaluation is stored together with the assignment. Method according to one of the preceding claims, characterized in that when evaluating the sensor data in step b) a driver-specific pattern (33) with a classifier (32) of the machine learning is detected and the control signal in step c) depending on the recognized pattern (33 ) is produced. Motor vehicle (1) with at least one sensor (4, 5, 6, 7, 8, 9, 10, 11) and an evaluation unit, which is designed to carry out a method according to one of the preceding claims.

Images