DE102019214709B4 - Detection of hazardous noises from a host vehicle - Google Patents

Abstract

Method for detecting hazardous noises of a host vehicle with the following steps: - Detecting (St1) noises emitted by a tire by means of a microphone; - Analyzing (St2) the noises with regard to whether an acceleration acting on the tire has a frictional connection limit of the tire exceeds; whereby it is checked whether the noises are emitted by tires of the host vehicle or tires of a third party vehicle.

Description

FIELD OF THE INVENTION

The present invention relates to a method for detecting hazardous noises from a host vehicle.

TECHNICAL BACKGROUND

The patent application DE 10 2018 200878 A1 shows a method for detecting dangerous noise in road traffic emitted by other vehicles.

the DE 10 2018 119 652 A1 relates to a vehicle having a chassis with at least one chassis component and a driver assistance system for detecting a state of the chassis component, the driver assistance system having an evaluation unit and a first acoustic sensor that records a sound generated by the chassis component when the vehicle is in motion and converts it to an electrical one Signal converts, and the evaluation unit uses the electrical signal to determine at least one piece of information about the state of the chassis component.

the DE 103 53 481 B4 discloses a method for determining a coefficient of friction between the roadway and the tires of a motor vehicle.

the DE 10 2004 044 788 B4 discloses a method for determining the coefficient of friction between a tire of a vehicle and a roadway.

SUMMARY OF THE INVENTION

Against this background, the invention is based on the task of recognizing further hazardous noises from vehicles.

According to the invention, this object is achieved by a method having the features of claim 1.

• - Ein Verfahren zum Detektieren von Gefahrengeräuschen eines Egofahrzeugs mit den folgenden Schritten: Erfassen von Geräuschen, die von einem Reifen emittiert wurden, mittels eines Mikrofons; Analysieren der Geräusche hinsichtlich dessen, ob eine auf den Reifen wirkende Beschleunigung eine Kraftschlussgrenze des Reifens übersteigt.

Accordingly, it is provided:

• A method for detecting hazardous noises in a host vehicle, comprising the following steps: detecting noises emitted by a tire by means of a microphone; Analyzing the noises with regard to whether an acceleration acting on the tire exceeds a frictional connection limit of the tire.

Vehicles within the meaning of this patent application are motor-driven land vehicles. Autonomous operation is the operation of devices, e.g. vehicles or mobile robots, without human intervention. In the case of vehicles, “autonomous operation” means that apart from determining a start or destination, no human intervention is required. In addition to fully autonomous vehicles (SAE level 5), there are also partially automated vehicles (SAE level 1-4). The invention is suitable for vehicles with a driver assistance function, that is to say fully autonomous or partially automated vehicles.

An interface is a device between at least two functional units at which an exchange of logical quantities, e.g. data, or physical quantities, e.g. electrical signals, takes place, either unidirectionally or bidirectionally. The exchange can be analog or digital. The exchange can also be wired or wireless.

A controller is an electronic module for controlling or regulating. Controls are used in all imaginable electronic areas in the automotive sector, as well as for controlling machines, systems and other technical processes.

An evaluation unit is a device that processes incoming information and outputs a result resulting from this processing. Electronic circuits such as central processing units or graphics processors are evaluation units.

Computer program products generally comprise a sequence of instructions which, when the program is loaded, cause the hardware to carry out a specific method that leads to a specific result.

An artificial neural network (ANN) is in particular a network of networked artificial neurons, which is simulated in a computer program. The artificial neurons are typically arranged on different layers. The artificial neural network usually comprises an input layer and an output layer, the neuron output of which is the only one of the artificial neural network that is visible. Layers between the input layer and the output layer are typically referred to as hidden layers. Typically, an architecture or topology of an artificial neural network is first initiated and then trained in a training phase for a special task or for several tasks in a training phase.

The term “topology of an ANN” encompasses all aspects relating to the structure of an ANN. This includes, for example, the number of Neurons of the ANN, the division of the neurons into the individual layers of the ANN, the number of layers of an ANN, the networking of the neurons and the weighting of the networking.

The training of the artificial neural network typically includes a change in the weight of a connection between two artificial neurons of the artificial neural network. The weight contains information on the strength of the consideration of an input of a neuron. The training of the artificial neural network can also include developing new connections between artificial neurons, deleting existing connections between artificial neurons, adapting threshold values of the artificial neurons and / or adding or deleting artificial neurons.

The reverse feeding includes the determination of a changed topology of the ANN, in particular of weights, in order to improve the determination of actual reaction signals based on the forward feeding.

A training system is a computing unit on which an ANN is trained.

In this application, training data are data pairs of input data that are to be processed by the ANN, as well as target result data that are to be determined by the ANN. During the training, the ANN is adapted on the basis of a comparison of target result data with the actual result data determined by the ANN, which results in a training effect.

The input data with which the ANN is fed in this application are noises or audio signals that encode noises. The input data contains hazardous noises, e.g. tire noises, tire squeals and the like.

An audio signal is an electrical signal that carries acoustic information.

An actual response signal can be derived from actual result data. A target reaction signal can be derived from target result data.

A microphone or microphone is a sound transducer that converts airborne sound as alternating sound pressure vibrations into corresponding electrical voltage changes as a microphone signal.

A directional microphone primarily picks up the frontal sound and thus has a directional characteristic. Sound from other directions is absorbed attenuated. The recorded sound is converted into electrical signals.

This application relates to various vehicles that are related to one another. Vehicles that use a method according to the invention are referred to below as ego vehicles. Vehicles that drive behind or in front of the ego vehicle, for example, are referred to as third-party vehicles.

The driving dynamics include the determination of distance, time, speed, acceleration, energy expenditure, heating of engines, drive forces, power, resistance to movement, trailer loads and vehicle efficiency. Driving dynamics uses technical, physical, mathematical and statistical principles. Driving dynamics considers the three translational movements in the direction of the main axes to be the spatial movement of bodies, namely the longitudinal movement along the longitudinal axis, the transverse movement along the transverse axis and the lifting movement along the vertical axis, often combined with the longitudinal movement when driving downhill or uphill. Lifting platforms and elevators implement a pure lifting movement. The driving dynamics also consider the three rotational movements around the three main axes, namely yawing around the vertical axis, pitching around the transverse axis and rolling around the longitudinal axis. Driving dynamics also consider two types of vibrations, namely translational and rotational vibrations.

The term driving dynamics control or electronically controlled stability, often also abbreviated to "ESP" for electronic stability program, describes an electronically controlled driver assistance system for vehicles that counteracts the vehicle breaking out by braking individual wheels.

As a connecting element between the vehicle and the roadway, a tire has a special function, as it supports the forces that occur in relation to the environment. Tire properties are influenced by operating conditions.

Wheel load, slip, camber and slip angle determine the size of the transmissible lateral force. The coefficient of friction between the tire and the road surface limits the forces that can be transmitted. These influencing variables essentially determine the transmission behavior of the tire.

The adhesion limit describes the maximum force that can be transmitted between the tire and the road surface. If the frictional connection limit is related to the normal force, a dimensionless one is obtained Coefficient that can be compared with the static friction coefficients of other friction pairings.

A sensor, also known as a detector, (measured variable or measuring) pick-up or (measuring) sensor, is a technical component that has certain physical, chemical properties or states, e.g. B. temperature, humidity, pressure, speed, brightness, acceleration, pH value, ionic strength, electrochemical potential and / or the material nature of its environment qualitatively or quantitatively as a measured variable. These variables are recorded using physical or chemical effects and converted as sensor data into an electrical signal that can be processed further. Vehicle sensors are mounted on a vehicle in order to detect a vehicle environment. Sensor data that are recorded by vehicle sensors are sensor data related to the vehicle environment.

The basic idea of the invention is to record tire noises of a host vehicle and to analyze the noises with regard to an acceleration acting on the tire.

If an acceleration acting on a tire exceeds a traction limit of a tire, this can lead to dangerous driving situations. Typically, accelerations of this kind which exceed a frictional connection limit of a tire cause typical noises, for example a tire screeching or the tire spinning.

This is advantageous for (partially) autonomous vehicles, for example, since they cannot “hear” such noises without the appropriate hardware and evaluation of sensor data.

It is thus possible to react to a detected acceleration noise or braking noise, which can be assigned to exceeding a frictional connection limit, by intervening in the driving dynamics of a vehicle in a controlling manner, for example. This can include, for example, measures to avoid accidents or to reduce the severity of an accident.

Advantageous refinements and developments emerge from the further subclaims and from the description with reference to the figures of the drawing.

According to a preferred development of the invention, a direction of origin and / or a volume of the noises is determined such that a tire or also a pair of tires that emits the noises can be inferred.

In this way, it is possible to intervene more specifically in a vehicle dynamics control to avoid accidents.

According to a preferred development of the invention, the acceleration acting on the tire is a transverse acceleration.

Accordingly, for example, noises can be detected which are emitted due to oversteering or due to understeering of a vehicle.

If, for example, oversteer is detected, a vehicle dynamics control system can be used to intervene in the vehicle control system in such a way that the vehicle is steered in the direction of the erupting rear end.

Such driving maneuvers are often unintuitive for drivers and require routine driving experience. Accordingly, the safety of a vehicle can be improved if corresponding driving maneuvers are initiated automatically.

Alternatively, it is also conceivable to detect noises which can be traced back to a longitudinal acceleration acting on the tire that exceeds a frictional connection limit.

According to a preferred development of the invention, a result of the analysis of the noises is validated by means of further sensors, for example acceleration sensors. Alternatively or additionally, it is also conceivable to validate further sensor data on the basis of the result of the analysis of the noises.

Further sensor data can thus be checked or the result of a method as described above can also be checked. The robustness of the method, as described above, or the robustness of further sensors is thus increased.

According to the invention, for example, a further method and / or further sensors are used to check whether the noises are emitted by tires of the host vehicle or by tires of a third-party vehicle. Accordingly, it can be provided that no interventions in the vehicle dynamics control are provided if it is detected that a tire noise was emitted by a third-party vehicle.

It can also be provided to check whether an emitted tire noise is in the direction of travel or is opposite to the direction of travel. Thus, for example, safe braking can be initiated if it is detected that a tire noise is coming from a direction of travel. Consequently possible collisions due to accident vehicles in the direction of travel can be avoided or reduced.

According to a preferred development of the invention, tire noises that reduce a frictional connection limit of a tire are also detected. Tire noises that reduce a frictional connection limit of a tire are, for example, splash noises due to wetness on a roadway, noises due to snow or ice or noises due to an unpaved roadway.

The computer program product according to one embodiment of the invention carries out the steps of a method according to the preceding description when the computer program product runs on a computer, in particular a computer internal to the vehicle. When the program in question is used on a computer, the computer program product produces an effect, namely the detection of an acceleration acting on a tire with an effect endangering the host vehicle.

Figure list

• 1 ein schematisches Blockdiagramm gemäß einer Ausführungsform der Erfindung;
• 2 ein schematisches Blockdiagramm eines Trainingssystems zum Trainieren eines künstlichen neuronalen Netzwerks gemäß einer Ausführungsform der Erfindung;
• 3 ein schematisches Blockdiagramm gemäß einer Ausführungsform der Erfindung;

The present invention is explained in more detail below with reference to the exemplary embodiments specified in the schematic figures of the drawings. It shows:

• 1 a schematic block diagram according to an embodiment of the invention;
• 2 a schematic block diagram of a training system for training an artificial neural network according to an embodiment of the invention;
• 3 a schematic block diagram according to an embodiment of the invention;

The accompanying drawings are intended to provide a further understanding of the embodiments of the invention. They illustrate embodiments and, in conjunction with the description, serve to explain principles and concepts of the invention. Other embodiments and many of the advantages mentioned emerge with a view to the drawings. The elements of the drawings are not necessarily shown to scale with one another.

In the figures of the drawings, elements, features and components that are the same, functionally identical and have the same effect - unless stated otherwise - are each provided with the same reference symbols.

DESCRIPTION OF EXEMPLARY EMBODIMENTS

1 FIG. 3 shows a schematic block diagram of a method for detecting hazardous noises of a host vehicle. In step St1, noises emitted by a tire are detected by means of a microphone. In step St2, noises are analyzed with regard to whether an acceleration acting on the tire exceeds a frictional connection limit.

2 Figure 3 shows a block diagram of a training system 10 for training an artificial neural network. The training system 10 includes an interface 12th as well as an evaluation unit 20th with an KNN 22nd . The KNN 22nd includes several neurons that are in 2 simplified with 108a-f are shown. In doing so, the neurons form 108a , b an input layer 102 who have favourited Neurons 108c , d, e form a hidden layer 104 and the neuron 108f forms an output layer 106 .

The neurons 108a , b the input layer 102 are through the interface 12th with an audio signal 16 fed forward. The audio signal 16 is in the neurons 108a , b of the input layer weighted with initial weights. It can be provided that the audio signal 16 is broken down into several signal components and the signal components are weighted. It can be provided that one or more functions are applied to the weighted input data. The evaluation of the function forms the output value of a neuron 108a , b, attached to the neurons 108c , d, e of the next lower layer, i.e. the hidden layer 104 , are transferred as input values. It can be provided that the hidden layer 104 has multiple layers.

Similar to the entry layer 102 are sent to the neurons 108c , d, e weighted input values passed to the hidden layer and one or more functions are applied to the weighted input values. The evaluation of the functions that is applied to the weighted input values form the output values of the neurons 108c , d, e. These output values are sent to the neurons of the output layer 106 transferred as input values. In 2 are the neurons of the output layer 106 exemplified as a neuron 108f shown. The neuron 108f calculated from the input values received from the neurons 108c , d, e of the hidden layer 104 were passed, weighting the input values and applying one or more functions to the weighted input values, an output value. An actual reaction signal can be derived from this initial value 24 derive. This process is also referred to as feeding an ANN forward.

In a next step, the actual response signal 24 with the target response signal 18th , which of the evaluation unit 20th through the interface 12th was passed, compared.

The next step is the topology of the individual layers 102 , 104 , 106 of the KNN 22nd adapted in such a way that the KNN 22nd for the transferred audio signal 16 the target response signal 18th calculated. The adaptation of the topology 26th can include changing the weights, adding connections between neurons, removing connections between neurons and / or changing functions that are applied to weighted input values. This process is also referred to as reverse feeding of an ANN.

the 3 FIG. 3 shows a block diagram of a method for training an ANN according to an embodiment of the invention. The method comprises the steps S1-S4 . In the step S1 becomes a pair of signals from an audio signal 16 and a target response signal 18th , for example with regard to an adaptation of the driving dynamics provided. In the step S2 becomes the ANN 22nd with the audio signal 16 fed forward. In the step S3 becomes an actual response signal 24 due to feed forward S2 determined. In the step S4 becomes the artificial neural network 22nd due to a difference between the actual response signal 24 and the target response signal 18th fed backwards. There is a changed topology 26th of the ANN, in particular weights, in order to improve the determination of actual reaction signals based on the feed forward.

List of reference symbols

10

Training system

12th

interface

14th

Training data

16

Audio signal

18th

Target response signal

20th

Evaluation unit

22nd

artificial neural network

24

Actual response signal

26th

topology

102

Entry layer

104

hidden layer

106

Output layer

108a-f

Neurons

S1-S4

Procedural steps

St1-St2

Procedural steps

Claims (8)

Method for detecting hazardous noises from a host vehicle with the following steps: - Detecting (St1) of noises emitted by a tire by means of a microphone; - Analyzing (St2) the noises with regard to whether an acceleration acting on the tire exceeds a frictional connection limit of the tire; whereby it is checked whether the noises are emitted by tires of the host vehicle or tires of a third party vehicle. Procedure according to Claim 1 , with a tire being determined based on an original direction and / or volume of the noises which is experiencing an acceleration that exceeds its frictional connection limit. Method according to one of the preceding claims, wherein it is detected whether the acceleration acting on the tire is a transverse acceleration. Method according to one of the preceding claims, wherein a result of the analysis of the noises is validated by means of further sensors, in particular by means of an acceleration sensor, and / or further sensor data are validated on the basis of the result of the analysis of the noises. Method according to one of the preceding claims, wherein tire noises which reduce a frictional connection limit of a tire are furthermore detected. Computer program product with program code means in order to carry out the method according to one of the preceding claims. Method for training an artificial neural network (22) of a microphone system which is set up, the method according to one of the preceding Claims 1 - 5 perform, with the following steps: - providing (S1) at least one pair of an audio signal and a target response signal (18); - Forward feeding (S2) of the artificial neural network with the at least one audio signal (16); - Determination (S3) of an actual reaction signal (24) based on the forward feed (S2); - Backward feeding (S4) of the artificial neural network based on a difference between the actual reaction signal (24) and the target reaction signal (18), the audio signals having noises of a tire that can be assigned to an acceleration that exceeds the traction limit of the tire. Microphone system for detecting hazardous noises from a host vehicle according to a method according to one of the Claims 1 - 5 , with at least one microphone for detecting tire noises and an evaluation unit (20) which is set up according to the method according to Claim 8 to feed a trained artificial neural network (22) with the tire noises in order to assign a reaction signal (24) to the tire noises.

Images