GB2559141A

GB2559141A - A system and method for monitoring and detecting tyre wear

Info

Publication number: GB2559141A
Application number: GB1701287.3A
Authority: GB
Inventors: Josh Chiwoko Kudzayi
Original assignee: Jaguar Land Rover Ltd
Current assignee: Jaguar Land Rover Ltd
Priority date: 2017-01-26
Filing date: 2017-01-26
Publication date: 2018-08-01
Anticipated expiration: 2037-01-26
Also published as: GB201701287D0; GB2559141B

Abstract

A method and system for determining the degree of tyre wear in a vehicle (100, see fig 1b), comprising: a processing means for obtaing one or more date sets containing acoustic data relating to tyre noise associated with one or more parameters; detecting acoustic tyre noise with an audio sensor from one or more tyres on the vehicle; a machine learning process for determining a degree of wear and an output module for providing an output. The output may comprise a signal indicative of the determined degree of tyre wear. The system may comprise a vehicle control system for controlling one or more driving parameters, such as speed, brake, throttle or steering. The machine learning process may be configured to learn an acoustic response for a predetermined type of tyre or predetermined conditions. The machine learning process may also comprises a neural network with an input layer, a hidden layer and an output layer. Finally, the audio sensor may be a shotgun microphone.

Description

(54) Title of the Invention: A system and method for monitoring and detecting tyre wear

Abstract Title: System and method for monitoring and detecting tyre wear using an audio sensor (57) A method and system for determining the degree of tyre wear in a vehicle (100, see fig 1 b), comprising: a processing means for obtaing one or more date sets containing acoustic data relating to tyre noise associated with one or more parameters; detecting acoustic tyre noise with an audio sensor from one or more tyres on the vehicle; a machine learning process for determining a degree of wear and an output module for providing an output. The output may comprise a signal indicative of the determined degree of tyre wear. The system may comprise a vehicle control system for controlling one or more driving parameters, such as speed, brake, throttle or steering. The machine learning process may be configured to learn an acoustic response for a predetermined type of tyre or predetermined conditions. The machine learning process may also comprises a neural network with an input layer, a hidden layer and an output layer. Finally, the audio sensor may be a shotgun microphone.

502

Figure 5

1/10

100

Figure 1

112

2/10

Figure la

3/10

Figure lb

4/10

300

306

308

310

312

Figure 3

5/10

Omnidirectional Cardioid Hypercardioid Shotgun

Figure 2

A-weighted SPL [dB]

Figure 4

6/10

502

Figure 5

522

Figure 5a

7/10

504

522'

Figure 5b

8/10

Figure 6

9/10

Tyre Wear

701

700

Figure 7

10/10

800

802 \ X )

M Determine acoustic data sets for tyre wear

812

Indicate state of tyre to user

Change tyre if possible

804.

806.

814

Verify data sets

Measure tyre acoustic data

Determine degree of tyre wear v⁸¹⁰r

Invoke VCU

816 c

Compare tyre data with data sets

Adjust/control driving parameters

818

808 ~T

Figure 8

A SYSTEM AND METHOD FOR MONITORING AND DETECTING TYRE WEAR

TECHNICAL FIELD

The invention relates to a system and method for monitoring and detecting tyre wear. In particular, but not exclusively, the invention relates to an acoustic system and method for monitoring and detecting tyre wear in a vehicle which is capable of adjusting one or more driving parameters of the vehicle.

BACKGROUND

A vehicle tyre has a tread which is formed on the surface of the tyre having one or more grooves which define a tread pattern. The tread pattern and depth of the grooves are designed to ensure that the tyre makes an optimal contact with and thus gripping. The tread pattern also serves to clear standing water from between the tyre contact patch and the road surface.

Tyres are vulnerable to wear and sometimes damage and are often poorly maintained or their condition is overlooked by users. As a result, tyres may become worn with use and it is necessary to change one or more tyres when the tread is below an acceptable level. Typically the tread depth of the tyre degenerates with time and the fact that it has degenerated is generally detected visually when the tyre is checked by a user. Tyres are often checked sporadically.

A need thus exists to improve systems and methods for automatically monitoring tyre wear.

The present invention has been devised to mitigate or overcome at least some of the problems and disadvantages associated with the prior art.

SUMMARY OF THE INVENTION

Aspects and embodiments of the invention provide a tyre wear detection system, a vehicle and a method as claimed in the appended claims.

According to an aspect of the present invention, there is provided a tyre wear detection system for a vehicle, the system comprising: a processing means for obtaining one or more data sets containing acoustic data relating to tyre noise associated with one or more conditions; an audio sensor for detecting acoustic tyre noise from one or more tyres of the vehicle; a machine learning process for determining a degree of wear of the or each tyre in dependence on a comparison of the or each data set with the detected acoustic tyre noise; and an output module for providing an output in dependence on the determined degree of wear of the or each tyre.

In an embodiment, the output comprises a signal indicative of the determined degree of tyre wear.

In an embodiment, the tyre wear detection system comprises a vehicle control system for controlling one or more driving parameters in dependence on the determined degree of tyre wear.

In an embodiment, the vehicle control system is configured to control the one or more driving parameters to reduce a speed of the vehicle.

Optionally, the vehicle control system is configured to prevent further tyre wear.

In an embodiment, the or each driving parameter comprises a torque applied to one or more wheels of the vehicle.

In an embodiment, the or each driving parameter comprises at least one of a throttle control, a brake control and a steering assistance control.

In an embodiment, the machine learning process is configured to learn an acoustic response for a predetermined type of tyre in one or more predetermined conditions.

Optionally, the machine learning process is configured to learn an acoustic response for a predetermined type of tyre in a plurality of predetermined conditions.

In an embodiment, the machine learning process is configured to learn the acoustic response for one or more different types of tyre.

In an embodiment, the machine learning process is configured to learn the acoustic response for the one or more tyres for different types of terrain.

In an embodiment, the machine learning process is configured to learn the acoustic response for the one or more tyres for different types of road conditions.

In an embodiment, the machine learning process is configured to learn the acoustic response for the one or more tyres for different types of weather conditions.

In an embodiment, the machine learning process is configured to learn the acoustic response for the one or more tyres for different degrees of tyre wear.

In an embodiment, the machine learning process is configured to learn the acoustic response for the one or more tyres for one or more different inputs.

Optionally, the inputs include one or more of vehicle speed, wheel speed, engine and vehicle noise, factors associated with engagement of a terrain response mode and signals from the audio sensor.

In an embodiment, the machine learning process comprises a neural network.

Optionally, the neural network comprises an input layer; a hidden layer and an output layer.

Optionally, the or each layer includes one or more nodes and one or more connectors for connecting nodes.

Optionally, the input layer is configured to receive a plurality of inputs relating to the one or more input parameters for the machine learning process.

Optionally, the output layer is configured to generate a signal indicative of a determined degree of tyre wear.

Optionally, the hidden layer is configured to compare the or each data set and the detected acoustic tyre noise.

Optionally, the hidden layer is configured to apply a range of bias values to one or more nodes and/or connectors of the hidden layer until the neural network reaches a point at which the applied range of bias values gives a correct output.

Optionally, the audio sensor comprises a microphone and may comprise a shotgun microphone.

According to a further aspect of the invention, there is provided a vehicle comprising a tyre wear detection system of a first aspect of the present invention.

Optionally, vehicle control system and detection system may be integral or separate.

According to a further aspect of the invention, there is provided a method of determining a degree of tyre wear of one or more tyres of a vehicle, the method comprising: obtaining one or more data sets containing acoustic data relating to tyre noise associated with one or more parameters; detecting acoustic tyre noise from the one or more tyres of the vehicle; determining the degree of wear of the or each tyre based on a comparison of the or each data set with the detected tyre noise in a machine learning process; and providing an output in dependence on the determined degree of wear of the or each tyre.

Within the scope of this application it is expressly intended that the various aspects, embodiments, examples and alternatives set out in the preceding paragraphs, in the claims and/or in the following description and drawings, and in particular the individual features thereof, may be taken independently or in any combination. That is, all embodiments and/or features of any embodiment can be combined in any way and/or combination, unless such features are incompatible. The applicant reserves the right to change any originally filed claim or file any new claim accordingly, including the right to amend any originally filed claim to depend from and/or incorporate any feature of any other claim although not originally claimed in that manner.

BRIEF DESCRIPTION OF THE DRAWINGS

One or more embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

Figure 1 is a high level schematic diagram of an embodiment of a tyre wear monitoring system in accordance with the present invention;

Figure 1a	shows a diagram of the noise mechanism relating to tyre and road noise;
Figure 1b	shows a section through a rear wheel arch, showing the mounting of a microphone forming part of a tyre wear monitoring system in accordance with the present invention and its proximity to the wheel and tyre in the wheel arch;
Figure 2	is a schematic diagram of a number of different types of microphones which may be suitable for use in embodiments of a system according to the present invention;
Figure 3	is a block diagram of the tyre wear monitoring system of Figure 1;
Figure 4	is a graph showing frequency response for normal and slick tyres on a smooth and rough road surface;

Figures 5, 5a and 5b illustrate a recurrent neural network for use in an embodiment of a system according to the present invention;

Figure 6	is a high level schematic diagram of an embodiment of a vehicle control system, according to the present invention;
Figure 7	is a schematic diagram of further features of the vehicle control system, according to an embodiment of the present invention; and
Figure 8	is a flow diagram showing the method steps of a tyre monitoring method, according to an embodiment of the present invention.

DETAILED DESCRIPTION

A vehicle 100 is shown in Figure 1 having front wheels 102 and rear wheels 104 (only one of which is shown). The wheels each include tyres 106 having a tread 108. The vehicle 100 includes a wheel arch 110 around each wheel. At least one of the wheel arches is fitted with an audio sensor in the form of a microphone 112 as shown in the inset in Figure 1. In the illustrated embodiment, the microphone 112 is located in front of a rear wheel 104 of the vehicle. This position may be chosen to help protect the microphone 112 from objects that may impact against it. In embodiments, the microphone 112 may be positioned proximal to a front wheel 102, and may be located behind the wheel 102 to distance it from engine noise. Additionally or alternatively, an anti-wind foam (not shown) may be used to minimise wind turbulence effects. As will be appreciated, the location of the microphone 112 may be different in different circumstances. In another embodiment, the system may include a microphone situated in each wheel arch to check each tyre. Alternatively, only one tyre may be monitored and an assumption made that all the tyres wear at a similar rate. In an alternative embodiment, more than one tyre may be monitored by a single microphone if the directional beam pattern and range is appropriate for this.

It should be noted that the microphone 112 is one possible means of collecting or sensing noise generated in the vicinity of the tyre. Other audio sensing means or sensors are envisaged. For example, the audio sensor may comprise an audio transducer, receiver or any other appropriate device or system.

The microphone 112 is operable to detect sounds associated with the noise generated by the tyre being in contact with the road surface. In some embodiments, the microphone 112 may be positioned in front or to the rear of the wheel, between an angle of inclination of between about 30° and 60° relative to a surface on which the vehicle is travelling. This position may be chosen to make use of a horn effect of the generated noise which acts to enhance the signal to noise ratio in the detected noise and improve the immunity to interference from other sources of noise.

Figure 1a shows a diagram of the noise mechanism relating to tyre and road noise. A tyre 106 driving on a road 120, in a direction of rotation 122 may generate at least the following types of noise: surface vibrations 124; horn effect 126; beat of tread blocks 128; snap out of tread blocks 130; stick-slip 132; air pumping 134; groove resonance 136; macro roughness noises 138; mega roughness noises 140 and road texture impact 142. The noise from these tyre/road interactions is amplified, resulting in the microphone picking up more of the desired signal compared to background noise.

Figure 1b shows a cross-section through a rear side portion of a vehicle 100, showing the proximity of a microphone 112 in a wheel arch 110. In the figure, the microphone 112 is shown secured to the vehicle body structure via a bracket 144 and mounting screws 146, although other forms of mounting means, such as clips, are envisaged. The microphone 112 may be housed in a sill structure of the side of the vehicle body, or may be secured to the wheel arch 110 or wheel house structure as may be desirable for packaging and assembly for a given vehicle structure. Figure 1b shows the microphone 112 directed generally towards the lower region of the tyre, where the tyre 106 is in contact with the road in normal driving conditions. It will be appreciated that, depending on the directional nature of the microphone used and the range of vertical travel available between the tyre and the vehicle wheel arch, the location of the microphone may need to be tuned for a given vehicle application to optimise microphone performance in normal use.

Figure 2 shows a number of different types of microphone 200, 202, 204, 206 which may be suitable for use in embodiments of the present invention, depending on the location of the microphone 200, 202, 204, 206 relative to the vehicle 100. The different microphones include an omnidirectional microphone 200, a cardioid microphone 202, a hyper-cardioid microphone 204 and a shotgun microphone 206. A shotgun microphone 206 may be used positioned within a wheel arch behind a wheel of the vehicle 100. The shotgun microphone is a highly directional microphone with an elongated barrel which can be directed at a localized sound source. Other microphones may be used where the microphone has a different reception area and or when the microphone is located at a different locations. Where there is less than one microphone for each wheel arch, the microphone may be sampled at different times to receive audio from more than one tyre. In this case, the position of the microphone is chosen to provide a “line of sight” to each monitored tyre from a common location.

An overview of a system 300 according to an embodiment of the present invention is shown in Figure 3. The system 300 includes a microphone module 302. The microphone module 302 is connected to the microphone (not shown). The microphone module 302 receives audio signals from the microphone and carries out audio processing thereon to produce an audio output 304. The signal may be amplified, as required. The audio output 304 is processed by an analogue to digital converter (ADC) 306 to produce a digital audio signal. The output signal 304 is processed to extract the frequency response, the overall sound pressure, and psychoacoustic metrics which are then used as inputs for a machine learning process in the form of a neural network 308. It will be appreciated that the neural network 308 may be implemented as a software block or module within the system 300, either on a separate hardware/firmware component or partitioned into a software processing area, for example. The digital audio signal is processed via the neural network 308 to determine the state of the tyre as will be described in detail below.

The neural network 308 is configured to process the digital audio signal to determine and output the degree of wear of one or more tyres of the vehicle. The output is sent to a vehicle control unit (VCU) 310, as shown. The VCU 310 is configured to control and/or adjust one or more driving parameters 312 of the vehicle in dependence on the determined state of one or more of the tyres, as is described in detail below. Additionally or alternatively, the determination of the degree of tyre wear may be output to a user of the vehicle.

Figure 4 shows a graph of frequency response for normal and slick tyres on a smooth and rough road surface. The graph shows a weighted Sound Pressure level (SPL) in decibels against a frequency in Hertz. When used herein and throughout the specification, the term ’’normal tyres” is intended to cover tyres which have tread levels which are “normal” or greater than or equal to a predetermined threshold level. The threshold may be set at the legal minimum tread depth for a particularly country - e.g. 1,6mm for the UK, and/or it may be configured to give advanced warning to the driver of approaching minimum tread depth, and so may be set at a suitable margin above the legal minimum, e.g. 2mm. Similarly, the term “slick tyres” is intended to cover tyres with low levels of tread, which may be less than or equal to a predetermined threshold level, and which are used to model the responses of tyres which have insufficient tread. It will be appreciated that instead of using slick tyres to model worn tyres, measurements could be taken from worn tyres instead.

In an embodiment, a frequency response is measured for one or more tyres on the vehicle. The measurements can then be compared with the known responses of normal and slick tyres, such as those shown in Figure 4. From a comparison of measured readings with the known responses it is possible to determine if the tyre is behaving like a normal tyre or like a slick tyre. It is possible to make a determination of degree of tyre wear from detected frequency response alone, particularly for smooth surfaces. However, the use of a neural network provides a more robust determination of the degree of tyre wear.

In order to determine whether the measured responses of a tyre are above or below a predetermined threshold, the present invention makes use of a deep learning technique based on a recurrent neural network. This will now be described in detail with reference to Figures 5, 5a and 5b.

The neural network 500 includes an input layer 502, a hidden layer 504 and an output layer 506. The neural network 500 receives a plurality of inputs 508, 510, 512, 514, 516 from various sources. These include vehicle speed 508, wheel speed 510, engine and vehicle noise 512, factors associated with the engagement of terrain response selected mode 514 and signals 516 from one or more audio sensors (which may comprise the microphone 112).

The output layer 506 includes a first output 518 which indicates that the tyre is normal and a second output 520 which indicates that the tyre is worn. In response to the first output 518, the operational state of the vehicle may be maintained, for example. In response to the second output 520, one or more driving parameters could be adjusted accordingly. Additionally or alternatively, the outputs 518, 520 may be communicated in an appropriate manner which enables the driver or the vehicle control system to take any necessary actions. The outputs may be displayed at an appropriate location on an instrument panel (not shown) of the vehicle, for example.

In some embodiments a third output (not shown) may be provided which indicates that the tyre is becoming worn. In response to this output, one or more driving parameters could be adjusted to optimise vehicle characteristics for the tyres in their present condition so as to delay or mitigate further damage to the tyres.

Figures 5a and 5b illustrate a further configuration of a neural network 500 for use in embodiments of the present invention. As can be seen in Figure 5a, the neural network 500 may comprise multiple hidden layers 504. It will be understood that any number of hidden layers 504 may be implemented. In addition, the number of input, output and hidden nodes is different from the embodiment shown in Figure 5. Each node 522 is linked to each node 522 in the next layer by connectors 524. Referring to Figure 5b, the connectors and the nodes in the hidden layer are assigned a bias which the system will compute during a training process and which are then validated during a testing process. This will allow system to be given new data and still be capable to determine tyre wear regardless of which surface the vehicle is driving on, the tyre compound, tyre pressure and tyre size.

The training and testing processes will now be described with reference to Figures 5, 5a and 5b.

During training, a plurality of data sets relating to the frequency response of a tyre in a variety of different changeable conditions are collected. The conditions may include different surfaces on which the vehicle is travelling; different speeds of the vehicle; vehicle engine speed; different tyre size; different tyre pressure; different tyre compounds, for example. For each combination of changeable conditions a data set is produced which is associated with the conditions. For each data set, a frequency response of both a “normal” and “slick” tyres is obtained and associated therewith.

More specifically, the system is trained by entering the various data sets associated with the known tyre conditions into the neural network 500 as inputs 508, 510, 512, 514, 516 for the input layer 502. A measured frequency response of a tyre is then obtained. This may be obtained using the microphone 112, as shown in Figure 1. This frequency response is also entered into the neural network 500, which is used to predict if the data obtained is indicative of a worn tyre or a normal tyre. In an embodiment, the neural network 500 may be configured to have specific pre-set bias values for different types of tyre or intrinsic tyre conditions, as shown in Figure 5b.

After each prediction, the system will check if the prediction is correct. Since the data entered belongs to known conditions, it will be known whether the output from the neural network 500 is correct. During the training process, the bias applied to the nodes and connectors in the hidden layer 504 may be adjusted to ensure a correct output is given by the neural network. For example, the system may try a number of combinations of bias values in the hidden layer 504 until the neural network 500 reaches a point at which the bias gives a correct output. In other words, until the system accurately identifies the degree of wear of the tyre. To further refine the neural network 500, data sets associated with a wide range of tyre compounds, tyre wear level, and different road surfaces are entered into the neural network 500.

In an embodiment, the neural network 500 may be implemented as a recurrent neural network, which has the benefit of using the output layer 506 as an input to the hidden layer 504. This means that once the system knows it has worn tyres it will use this condition as an input parameter, thereby enabling the system to predict level of wear to a better extent. In addition, through the use of the bias associated with nodes and connectors of the neural network 500, the system may distinguish all inputs and may learn the relation between the inputs and outputs i.e. certain frequency responses are present in certain terrain conditions.

As discussed above, an object of the present invention is to determine the condition of a tyre so that if the tyre is worn or the tread had reduced below a preferred level, the driver is notified and thus the tyre may be changed at the earliest opportunity. However, in some instances it is not always possible to change the tyre straight away. If this is the case, it may be possible to adjust one or more driving parameters of the vehicle in order to prevent or mitigate further damage to the tyre. The driving parameter adjustments may be implemented using the VCU 310, as discussed above. Additionally or alternatively, the state of the tyres may be indicated by means of a visual and/or audible indicator to a user of the vehicle. The indicator may be a simple indicator, such as “good or bad”, “normal or worn”, for example, or may be indicated in a more qualitative manner by indicating a degree of wear.

Referring to Figure 6, the VCU 310 includes a speed progress (SP) control system and a stability control system (SCS) 602, both being known components of existing vehicle control systems. The SCS 602 is configured to detect and reduce a loss of traction. For example, when a loss of steering control is detected, the SCS may automatically apply one or more brakes of the vehicle to steer the vehicle in a desired direction.

The VCU is connected to a prime mover, such as an engine 604, and provides signals thereto and receives signals therefrom. It should be appreciated that the engine may be an internal combustion engine, but more broadly, should be interpreted to cover any vehicle power unit such as an electric motor, a hybrid power system and the like, as will be readily understood. During driving, positive drive torque is applied to the wheels via a vehicle powertrain comprising a prime mover, such as an engine 604, a VCU 310 and means to transmit the torque from the engine to the wheels, such as a transmission. The transmission may comprise a gearbox and the VCU 310 may be in electrical communication with the transmission in addition to the engine 604 so as to be able to coordinate the operation of the engine and transmission together as will be described below.

In addition, the VCU 310 is in communication with a gearbox 606, a braking system 608 and an accelerator 610. Although not shown in detail in Figure 6, the VCU 310 may further include a Dynamic Stability Control (DSC) function block, a Traction

Control (TC) function block, an Anti-Lock Braking System (ABS) function block and a Hill Descent Control (HDC) function block. These function blocks provide outputs indicative of, for example, DSC activity, TC activity, ABS activity, brake interventions on individual wheels and engine torque reduction requests from the VCU 310 to the engine 604.

The VCU 310 is also in communication with a tyre wear detection system 612 which is operable to determine whether or not the tyre or tyres are worn. The detection system 612 may be equivalent to system 300 described in detail above. The tyre wear detection system 612 is provided with a tyre wear detection HMI 614 through which the tyre wear detection system 612 can indicate the state of the tyres to the user of the vehicle.

The tyre wear detection system 612 is also configured to adjust or control one or more vehicle parameters, such as vehicle speed, brake response, throttle response or steering response, for example. The adjustments in these parameters may be implemented via one or more of a, gearbox 606, braking system 608, traction control system, stability management system, anti-lock braking system, assisted steering system, etc. to maintain stability and optimise traction for the user, as will be understood.

In this way, the system is operable to warn a user of the vehicle of tyre condition and/or to adjust vehicle parameters to ensure a maximum level of traction.

Figure 7 illustrates the means by which the vehicle powertrain is controlled in a tyre wear detection system 700. The tyre wear detection system 700 may be equivalent to the system 612 or system 300 shown in the preceding Figures. A vehicle speed sensor (not shown) provides a signal indicative of a target vehicle speed 702. The target vehicle speed 702 may be a maximum suitable speed set by the system 700 in dependence on the determined degree of tyre wear 701, or may be selected by a user of the vehicle in the event that the target speed selected by the user is less than the defined maximum suitable speed. The system 700 includes a comparator 704 which compares the target speed 702 with a measured reference speed 706 and provides an output signal 708 indicative of the comparison. The output signal 708 is provided to a processing unit 710 of the VCU which interprets the output signal 708 as either a demand for additional torque to be applied to the vehicle wheels, or for a reduction in torque to be applied to the vehicle wheels.

An output 712 from the processing unit 710 is provided to the driveline for the vehicle wheels 714, 716 so as to either increase or decrease the torque applied to the wheels, depending on whether there is a positive or negative demand for torque from the processing unit 710. In order to initiate the necessary positive or negative torque being applied to the wheels, the unit 710 may either command that additional power is applied to the vehicle wheels or that a braking force is applied to the vehicle wheels, either or both of which may be used to implement the change in torque that is necessary to maintain the vehicle at a given speed, or to increase or decrease the speed of the vehicle to a given speed, in dependence on the state of the tyres. The state of the tyres can be derived by the system by a tyre wear indicator signal indicative of the degree of tyre wear. In the illustrated embodiment, the positive drive torque applied to the vehicle wheels is controlled individually so as to maintain the target vehicle speed, but in another embodiment the torque applied to the wheels may be controlled collectively to maintain the target speed. Additionally or alternatively, the rate of change of positive drive torque provided to the wheels by the vehicle powertrain may also be controlled by the VCU 310 in dependence on the determined tyre wear condition of the vehicle tyres.

The vehicle may also be provided with additional sensors (not shown) which are representative of a variety of different parameters associated with vehicle motion and status. The signals from the sensors provide, or are used to calculate, a plurality of driving condition indicators (also referred to as terrain indicators) which are indicative of the nature of the terrain conditions in which the vehicle is travelling. The signals are provided to the VCU 310 which determines the most appropriate control mode for the various subsystems on the basis of the terrain indicators, and automatically controls the subsystems accordingly.

The sensors (not shown) on the vehicle may include, but are not limited to, one or more of the following: wheel speed sensors, as mentioned previously and as shown in Figure 7, an ambient temperature sensor, an atmospheric pressure sensor, tyre pressure sensors, sensors to detect yaw, roll and pitch of the vehicle, a vehicle speed sensor, a longitudinal acceleration sensor, an engine torque sensor (or engine torque estimator), a steering angle sensor, a steering wheel speed sensor, a gradient sensor (or gradient estimator), a lateral acceleration sensor on the stability control system (SCS), a wheel speed sensor, a road roughness sensor, a ride height sensor, a brake pedal position sensor, an acceleration pedal position sensor and longitudinal, lateral, vertical motion sensors, a wading event detection sensor, a rain sensor, a vehicle proximity sensor.

The VCU 310 may evaluate the various sensor inputs to determine the probability that each of a plurality of different control modes for the vehicle subsystems is appropriate, with each control mode corresponding to a particular road surface type (for example, concrete, asphalt etc. and whose surface, with which the tyre is in contact, is determined as being wet or dry). Additionally or alternatively, the VCU 310 may use the various sensor inputs to determine the probability that each of a plurality of different control modes for the vehicle subsystems is appropriate, with each control mode corresponding to a particular the terrain type over which the vehicle is travelling (for example, mud and ruts, grass, gravel, snow). The VCU 310 then selects which of the control modes is most appropriate and controls various vehicle parameters accordingly. In an embodiment, if the VCU 310 determines that the vehicle 100 is performing a wading event, where the vehicle is traversing deep water such as crossing a ford, the system 300 may suspend active monitoring of tyre wear until a pre-determined time after completion of the wading event. In this way, sounds of the water splashing around the tyre and wheel arch during a wading event, picked up by the microphone 112, will not result in erroneous tyre wear estimates and false tyre wear notifications being issued to the driver.

The nature of the terrain over which the vehicle is travelling may also be utilised in the control system 700 to determine an appropriate increase or decrease in drive torque to be applied to the vehicle wheels based on the tyre wear. For example, if the driver selects a target speed that is not suitable for the nature of the terrain over which the vehicle is travelling, the system may be operable to automatically reduce the vehicle speed. In some cases, for example, the driver selected speed may not be achievable over certain terrain types, particularly in the case of uneven or rough surfaces. If the system selects a speed that differs from the driver-selected target speed, a visual indication of the speed constraint is provided to the driver via the HMI 614.

Referring to Figure 8, a set of steps for a method 800, according to an embodiment of the invention, is now described. In a first step 802 one or more acoustic data sets are determined for a given tyre on a given surface. Different data sets are collected for all types of tyre, terrain and any other parameter relevant to the noise generated by a tyre on a surface.

In step 804, the data sets are verified using a neural network and verified and checked data sets are stored for future use. In step 806, data is collected from one or more of the tyres using the system described above. In step 808 measured tyre data is compared with the data sets. This is also carried out within the neural network, where various inputs are received, as described above, and the neural network can determine if the tyres are worn or normal. In step 810 the degree of tyre wear is determined. The state of the tyre is then indicated to the user in step 812 or sent directly to the VCU. The indication may be on the instrument panel and may include an alarm or any other type of warning mechanism to provide suitable notification to the vehicle driver. The indication may also advise of actions which should take place as a result of the degree of tyre wear. The indication to the user may also be that a vehicle control system has been implemented to control further driving parameters until the tyre is repaired or replaced. In extreme conditions the car may be immobilised.

If possible, at step 814 the tyre or tyres may be changed. If this is not possible the tyre wear data may be communicated to the VCU in step 816. As described above, the VCU may adjust one or more driving parameters to take into account the degree of tyre wear in step 818. The system is operable to adjust or control any parameters that can lead to break in traction of the tyre with the road, such as for example, throttle, steering and braking controls.

In an embodiment, the VCU may be operable to adjust the terrain response setting of the vehicle. Each terrain response setting will have different dynamic responses. Therefore, the terrain setting of the vehicle could be changed from “normal” to for example, Grass; Gravel; or Snow mode in the event that the degree of tyre wear is determined to be worn as these settings related to is a predefined setting when the vehicle is operating in slippery conditions. Additionally or alternatively, the settings could give rise to a change in throttle response to reduce the likelihood of wheel spin. Brake sensitivity may be reduced to stop a driver completely locking one or more of the wheels. Steering assistance, sensitivity and responsiveness may be adjusted so that small, unintentional steering inputs are reduced. The suspension settings may be adjusted to either soften or stiffen the suspension to maximise the available grip. Differential bias setting may be adjusted to avoid wheel spinning. Other changes may be made depending on the VCU control settings which exist in the vehicle. Where the vehicle is fitted with a terrain setting this may serve as an input for the neural network.

The above described method steps can be repeated as often as necessary. The 5 frequency of the monitoring may be augmented if there is an indication that the tyre tread is approaching a predetermined level. Additional steps may be taken as is necessary and as required by the circumstances.

References herein to a block such as a function block are to be understood to include 10 reference to software code for performing the function or action specified in which an output is provided responsive to one or more inputs. The code may be in the form of a software routine or function called by a main computer program, or may be code forming part of a flow of code not being a separate routine or function. Reference to function blocks is made for ease of explanation of the manner of operation of the present invention.

It will be understood that the embodiments described above are given by way of example only and are not intended to limit the invention, the scope of which is defined in the appended claims.

Claims

CLAIMS:

1. A tyre wear detection system for a vehicle, the system comprising:

a processing means for obtaining one or more data sets containing acoustic data relating to tyre noise associated with one or more conditions;

an audio sensor for detecting acoustic tyre noise from one or more tyres of the vehicle;

a machine learning process for determining a degree of wear of the or each tyre in dependence on a comparison of the or each data set with the detected acoustic tyre noise; and an output module for providing an output in dependence on the determined degree of wear of the or each tyre.

2. The tyre wear detection system as claimed in claim 1, wherein the output comprises a signal indicative of the determined degree of tyre wear.

3. The tyre wear detection system of claim 1 or 2, comprising a vehicle control system for controlling one or more driving parameters in dependence on the determined degree of tyre wear.

4. The tyre wear detection system as claimed in claim 3, wherein the vehicle control system is configured to control the one or more driving parameters to reduce a speed of the vehicle.

5. The tyre wear detection system as claimed in claim 3 or 4, wherein the or each driving parameter comprises a torque applied to one or more wheels of the vehicle.

6. The tyre wear detection system as claimed in any one of claims 3 or 5, wherein the or each driving parameter comprises at least one of a throttle control, a brake control and a steering assistance control.

7. The tyre wear detection system as claimed in any of claims 1 to 6, wherein the machine learning process is configured to learn an acoustic response for a predetermined type of tyre in one or more predetermined conditions.

8. The tyre wear detection system as claimed in claim 7, wherein the machine learning process is configured to learn the acoustic response for different types of tyre.

9. The tyre wear detection system as claimed in claim 7 or claim 8, wherein the machine learning process is configured to learn the acoustic response for the one or more tyres for different types of terrain.

10. The tyre wear detection system as claimed in any one of claims 7 to 9, wherein the machine learning process is configured to learn the acoustic response for the one or more tyres for different types of road conditions and or weather conditions.

11. The tyre wear detection system as claimed in any one of claims 7 to 10, wherein the machine learning process is configured to learn the acoustic response for the one or more tyres for different degrees of tyre wear.

12. The tyre wear detection system as claimed in any one of claims 7 to 11, wherein the machine learning process is configured to learn the acoustic response for the one or more tyres for one or more different inputs.

13. The tyre wear detection system as claimed in claim 12, wherein the inputs include one or more of: vehicle speed, wheel speed, engine and vehicle noise, factors associated with engagement of a terrain response mode and signals from the audio sensor.

14. The tyre wear detection system as claimed in any one of claims 1 to 13, wherein the machine learning process comprises a neural network.

15. The tyre wear detection system as claimed in any one of claims 1 to 14, wherein the neural network comprises an input layer; a hidden layer and an output layer.

16. The tyre wear detection system as claimed in claim 15, wherein the or each layer includes one or more nodes and one or more connectors for connecting nodes.

17. The tyre wear detection system as claimed in claim 15 or claim 16, wherein the input layer is configured to receive a plurality of inputs relating to the one or more input parameters for the machine learning process.

18. The tyre wear detection system as claimed in any of claims 15 to 17, wherein the output layer is configured to generate a signal indicative of a determined degree of tyre wear.

19. The tyre wear detection system as claimed in any of claims 15 to 18, wherein the hidden layer is configured to compare the or each data set and the detected acoustic tyre noise.

20. The tyre wear detection system as claimed in claim 16 or any claim dependent thereon, wherein the hidden layer is configured to apply a range of bias values to one or more nodes and/or connectors of the hidden layer until the neural network reaches a point at which the applied range of bias values gives a correct output.

21. The tyre wear detection system as claimed in any one of claims 1 to 20, wherein the audio sensor comprises a shotgun microphone.

22. A vehicle comprising a tyre wear detection system as claimed in any preceding claim.

23. A method of determining a degree of tyre wear of one or more tyres of a vehicle, the method comprising:

obtaining one or more data sets containing acoustic data relating to tyre noise associated with one or more parameters;

detecting acoustic tyre noise from the one or more tyres of the vehicle; determining the degree of wear of the or each tyre based on a comparison of the or each data set with the detected tyre noise in a machine learning process; and providing an output in dependence on the determined degree of wear of the or each tyre.

24. A tyre wear detection system substantially has hereinbefore described, with reference to the accompanying drawings.

25. A vehicle substantially has hereinbefore described, with reference to the accompanying drawings.

26. A tyre wear detection method substantially has hereinbefore described, with 5 reference to the accompanying drawings.

Application No: GB1701287.3 Examiner: Mr Robin Jones