FIELD

The invention relates to a monitoring system for a vehicle. BACKGROUND

The state of vehicle wheels is a significant factor in traffic safety, therefore it is important to monitor the state of wheels in order to prevent accidents and decrease wheel wear and fuel consumption.

In prior art solutions, the state of vehicle wheels has been monitored visually. There exist measuring systems where a measurement module is in- stalled in a vehicle wheel, and the measurement measures a property of the wheel, e.g. pressure and acceleration, and reports the measurements to a central processing unit (CPU) configured to process the measurements so as to determine whether a tire pressure has fallen in a wheel. There is, however, a need for improving the prior art systems. BRIEF DESCRIPTION

It is an object of the invention to provide a monitoring system for monitoring vehicle wheels and an apparatus allowing efficient monitoring of vehicle wheels. According to an aspect of the invention, the object is achieved with the subject matter of the independent claims. Embodiments are defined in the dependent claims.

LIST OF FIGURES

The invention will be described in greater detail with reference to the preferred embodiments and the accompanying drawings, in which

Figure 1 illustrates a first example of the structure of a monitoring system for vehicle wheels,

Figure 2 illustrates an example of the structure of a wireless measurement module,

Figure 3 illustrates an example of the structure of a central processing unit,

Figure 4 illustrates an example of vehicle wheel physics during wheel rotation,

Figure 5 illustrates a first example of acceleration of a vehicle wheel, Figure 6 illustrates a second example of acceleration of a vehicle wheel,

Figure 7 illustrates a flow diagram of a process for predicting future events related to vehicle wheels according to an embodiment of the invention, and

Figure 8 illustrates temporal regression models associated to measured properties of the vehicle wheels that may be used in the process of Figure 7.

DESCRIPTION OF EMBODIMENTS

With reference to Figure 1 , a monitoring system for vehicle wheels comprises measurement modules 104A, 104B to be attached to wheels 102A, 102B of a vehicle 100. The measurement modules may be wireless measurement modules 104A, 104B configured to communicate by means of a wireless communication signal 108A, 108B or, in another embodiment, they are config- ured to communicate by means of a wired communication signal. In an embodiment, the monitoring system also comprises a central processing unit (CU) 106, which may communicate with the measurement module 104A, 104B by means of the wired or wireless communication signal 108A, 108B. In the embodiment utilizing the wired communication signal, a wired signal connection may be provided between the measurement module and the CU. The embodiments are mainly described in connection with the wireless communications, but it does not limit the scope of the present invention to wireless solutions. The central processing unit 106 may be physically separated from the measurement modules, i.e. an individual device, or it may be comprised in one of the measurement modules 104A, 104B. The separate CU provides flexibility in the placement of the CU, e.g. to place it in a central place in the vehicle so that all the measurement modules are able to communicate with the CU. When the CU is integrated in one of the measurement modules, the number of physical devices in the system may be reduced.

The vehicle 100 may be a car, lorry, motorcycle and/or trailer, or any other automotive vehicle. However, the disclosed embodiments are not restricted to these examples and the vehicle may be any wheeled vehicle, the wheels 102A, 102B of which need to be monitored.

The vehicle wheel 102A, 102B typically comprises a rim and a tyre. A tyre is typically a gas-filled rubber structure. The central processing unit 106 may be installed in the vehicle 100, in which case the wireless central processing unit 106 may utilize the vehicle 100 structures, such as a user interface 120 and a power source. The user interface may comprise a display unit and/or a loudspeaker in a cockpit of the vehicle so as to display and/or playback information to a driver/operator.

In an embodiment, the monitoring system comprises a translational and/or a rotational acceleration sensor 1 10 for measuring translational acceleration of the vehicle 100. Translational acceleration measured from the rotational acceleration will be described in greater detail below.

With reference to a wireless measurement module 200 illustrated in

Figure 2, the wireless measurement module 200 typically comprises a communication unit (TRX) 206 of the measurement module with antennas 208, a state determination unit (SDU) 204 connected to the communication unit 206 of the measurement module, and an acceleration measurement unit (AMU) 202 connected to the state determination unit 204.

The acceleration measurement unit 202 determines acceleration information 214 associated with the motion of the wheel 102A, 102B in one or more directions. The acceleration information 214 is supplied to the state determination unit 204, which determines the value of a property of the vehicle wheel 102A, 102B by using the acceleration information 214.

In an embodiment, the measured property characterizes the number of revolutions that the vehicle wheel 102A, 102B has rotated within a determined time duration or a variable proportional to the number of revolutions, such as travelling distance.

In an embodiment, the measured property characterizes the radius of the vehicle wheel 102A, 102B or a variable proportional to the radius, such as the diameter of the wheel 102A, 102B or the length of the circumference. The radius of the wheel 102A, 102B or a property proportional to the radius can further be used for determining, for instance, the air pressure of the tyre of the wheel 102A, 102B, the amount of tyre tread, the load of the wheel 102A, 102B, how round the wheel is, the balance of the wheel, how well the wheel is fastened with bolts, and/or overall operational state of the wheel. Imperfect, i.e. not round, shape of the wheel typically shows as anomalies in a tangential acceleration, while poor balance and fastening may show in an existence of a lateral acceleration component. In an embodiment, the measured property characterizes an extreme condition of the wheel 102A, 102B, such as stress that exceeds a predetermined limit, which may be caused by the wheel 102A, 102B hitting an object or a roughness on the road surface, for example.

In an embodiment, the measured property characterizes lateral movement of the wheel 102A, 102B, which may result from a faulty bearing and/or breaking of the fastening of the wheel 102A, 102B. Such a phenomenon typically exhibits variance in the measured lateral acceleration, and such variance, e.g. lateral oscillation, may be detected. Lateral oscillation or vibra- tion may also indicate that fastening of the wheel is not appropriate, e.g. the bolts are loose. The lateral acceleration may also be used to detect wrong angles of the wheels. If the acceleration measurements show a steady (non- oscillating) lateral component, it may be an indicative of wrong wheel angle.

The measured property may be determined by using pre-known ref- erence parameters of the wheel 102A, 102B stored in a memory unit 212 and/or reference parameters determined by the state determination unit 204, such as reference radius, reference temperature, reference pressure and/or reference load of the wheel 102A, 102B.

The state determination unit 204 may be implemented, for instance, by means a computer program 218 executed by a digital processor 2 0 and stored in the memory unit 212 (MEM).

The digital processor 210 may download the computer program 218 from the memory unit 212, the computer program 218 including encoded instructions for executing a computer process in the digital processor 210. The computer process may execute an algorithm for determining the value of the measured property characterizing the state or characteristic of the vehicle wheel 102A, 102B.

As an element for measuring the acceleration, the acceleration measurement unit 202 may comprise, for example, an acceleration sensor which is piezoelectric or implemented by other known technology. The structure and function of acceleration sensors are known per se to a person skilled in the art, therefore they will not be discussed in greater detail herein.

The digital processor 210 supplies the measured property 216 it has determined to the module communication unit 206, which communicates the measured property 216 by means of a wireless communication signal 220 to the central processing unit 106, for instance. The wireless communication signal 220 may be a radio frequency signal, in which case the module communication unit 206 typically comprises a radio transmitter.

In an embodiment, the radio transmitter is based on Bluetooth tech- nology.

In another embodiment, the radio transmitter implements a wireless local area network (WLAN, Wireless Local Access Network) (WPAN, Wireless Personal Area Network), which may be based on the IEEE 802.15.4 protocol on a 2.4 GHz band, for example. In an embodiment, the radio transmitter im- plements a ZigBee interface.

In an embodiment, the wireless communication signal 220 is based on the use of a magnetic component of the electromagnetic field. In such a case the antenna 208 may be replaced with an induction coil. The oscillating frequency of the electromagnetic field may be hundreds of kilohertz, for exam- pie, but the disclosed solution is not restricted to this frequency range.

In an embodiment, the wireless measurement module 200 comprises a tag memory (TM) 222 for storing the property the state determination unit 204 has determined. The tag memory 222 may be implemented, for instance, by means of a memory unit 212 or some other element capable of storing data. The tag memory 222 may be connected, for instance, by means of the digital processor 210 to the module communication unit 206, whereby data contained in the tag memory 222 may be transferred by means of the wireless communication signal 220 to the central processing unit 106, for example. The tag memory 222 may also include reference parameter values of the wheel as well as other information on the wheel 102A, 102B, such as the tyre type.

With reference to the example of Figure 3, a central processing unit 300 comprises an antenna 308, a communication unit (TRX) 302 of the central processing unit, connected to the antenna 308, a processor (PU) 304 connected to the communication unit 302 of the central processing unit, a memory unit 310 connected to the processor 304, a translational acceleration sensor (TAC) 306 connected to the processor 304, and a user interface (Ul) 314 connected to the processor 304. The user interface may comprise a display in a cockpit of the vehicle so that the processor may display information to a driver while the vehicle is moving. . The user interface, the communication module, and the processor may be combined into a single casing, some of them may be comprised in the same casing while one component is in another casing, or they all may be comprised in different casings but able to communicate with each other. For example, when the processor and the communication unit are comprised in one of the measurement modules in a wheel, the user interface may be comprised in the cockpit.

The communication unit 302 of the central processing unit may implement a wireless interface based on BlueTooth, WLAN, WPAN, ZigBee or magnetic data transfer, for example.

The communication unit 302 of the central processing unit communicates the wireless communication signal 220 with the wireless measure- ment module 200 of Figure 2 by receiving the communication signal 220 via the antenna 308. The communication unit 302 of the central processing unit supplies the measurement results included in the wireless communication signal 220 to the processing unit 304.

The processing unit 304 processes the measurement results on the basis of encoded instructions 312 stored in the memory unit 310.

In an embodiment, the central processing unit 300 is a computer, in which the communication unit 302 of the central processing unit may be, for example, a network card which implements a Bluetooth connection of a wireless local area network. The computer may be, for example, a portable com- puter (laptop) or a desktop computer (PC). The computer may be placed in the facilities of a wheel service station, for example, in which case the computer may read data included in the tag memory 222 of the wheel 102A, 102B, such as driving kilometres, the number of revolutions, and/or the property associated with an extreme condition.

In an embodiment, the central processing unit 300 is a mobile phone, a PDA device (Personal Digital Assistant) or another portable electronic device which implements a wireless interface needed to transfer the communication signal 220.

In an embodiment, the central processing unit 300 is a data pro- cessing device installed in the vehicle 100. The data processing device may also be connected to a CAN bus (CAN, Controller Area Network) of the vehicle, by which the data processing device may communicate with the vehicle systems.

With reference to the example of Figure 3, in an embodiment the monitoring system comprises a plurality of wireless measurement modules (MM#1 , MM#2, MM#N) 314A, 314B, 314C, which may have the same structure as the wireless measurement module 200 according to Figure 2. Each wireless measurement module 314A to 314C is configured to be attached to a different wheel 102A, 102B of the vehicle 100.

Each wireless measurement module 314A to 314C determines ac- celeration information of the wheel 102A, 102B, determines the measured value of a property characterizing a characteristic or state of the wheel 102A, 102B, includes the measured value in the wireless communication signal 316A to 316C and transmits the wireless communication signal 316A to 316C to the central processing unit 300. In another embodiment, the measurement module may process one or more of the measured values into another type of measurement result characterizing a characteristic or state of the wheel 102A, 102B. The measurement module may, for example, compute an air pressure in the tire or any other parameter that may be derived from the measurement results, and transmit the parameter as the measurement results to the central pro- cessing unit 300.

Each wireless communication signal 316A to 316C may have a measurement module -specific identifier, which may be implemented by means of the frequency or time frame of the wireless communication signal 316A to 316C or by means of the code included in the wireless communication signal 316A to 316C.

The communication unit 302 of the central processing unit receives the wireless communication signals 316A to 316C and supplies the parameters included in the wireless communication signals 316A to 316C to the processor 304 through an interface between the communication unit 302 and the proces- sor 304. The processor 304 may compare the measured property values received from different wireless measurement modules 314A to 314C and execute a predetermined function if the comparison fulfils predetermined conditions. The processor 304 may compare the measurement results associated with different wheels with each other and, upon detection that the measure- ment results of a given wheel differ from the measurement results of at least the other wheels, the processor 304 may initiate a function to report the detected deviation. For example, if the processor 304 detects that a measurement value is related to air pressure in the tyres and that a measurement values of a given wheel is lower than the measurement values of the other wheels by a determined amount defined by a preset threshold, for example, the processor may output a signal raising an alarm. The alarm may comprise lighting a warning light in the user interface of the vehicle.

In an embodiment, the processor 304 compares the measurement results to one or more reference values stored in the memory 310 beforehand. Upon detection of a difference above a predetermined threshold between the measurement results and the reference value(s), the processor may determine that the wheel is faulty. In another embodiment, the processor 304 may select one of the wheels as a reference wheel and compares measurement results related to other wheels to the measurement results of the reference wheel. Upon detection of a sufficient deviation, e.g. a difference above said predeter- mined threshold, between the measurement results of another wheel and the measurement results of the reference wheel, the processor 304 may determine which one of the tires is a faulty one by comparing the measurement results to the measurement results of a third wheel. A wheel whose measurements deviate from the measurements of the other two (or more) wheels by the deter- mined amount is determined to the faulty one, and the processor may indicate the faulty wheel in the report so as to facilitate the repair or replacement of the faulty wheel. The faulty wheel may be the reference wheel and, therefore, the processor may be configured to verify the faulty wheel by adding more wheels to the evaluation.

In an embodiment, the measured property characterizes the wheel's

102A, 102B rotational frequency, radius or variable proportional to the radius, such as the diameter or circumference length of the wheel 102A, 102B. The processing unit 304 may compare the measurement values of different wheels 102A, 102B and, in case of detecting a predetermined or dynamically updated deviation between the measurement results associated with said different wheels, indicate this deviation to the user by means of the user interface 314 of the central processing unit 300, for instance. A deviation in the rotational frequency of a wheel 102A, 102B from the rotational frequencies of other wheels may be an indication that the air pressure has diminished or the tyre tread has become smaller in the wheel 102A, 102B in question.

In an embodiment, the monitoring system is integrated into the anti- theft system for vehicle wheels, which utilizes acceleration information measured from the vehicle wheels 102A, 102B. In this case, the wireless measurement module 104A, 104B determines, for instance, kinetic state information characterizing the kinetic state of the wheel 102A, 102B on the basis of acceleration measurement, generates a wireless communication signal 108A, 108B on the basis of the kinetic state information and transmits the wireless communication signal 108A, 108B to the central processing unit 106. The central processing unit 106 receives the wireless communication signal 108A, 108B and may execute a predetermined alarm function if the wireless communication signal 108A, 108B fulfils the predetermined conditions.

With reference to Figure 4, let us examine rotational characteristics of a wheel 400 rotating on a base. The wheel comprises a tyre part 402 and a rim part 404.

In the example of Figure 4, a wireless measurement module 408 is placed at the distance of a measurement radius 410 from the centre point of the wheel 400. The radius of the wheel 400 is denoted by the reference number 406. The measurement radius 410 affects the measurement range required by the wireless measurement module 408. In the selection of the measurement radius 410, it is reasonable to take into account the maximum rota- tional speed of the wheel 400 and the dynamic operating range and error margin of the acceleration measurement unit 202. The measurement radius 410 may vary between 0 mm and 10 cm, for instance. In an embodiment measuring the radial acceleration, the measurement radius is preferably 0 mm (the centre of the wheel). However, the measurement radius above 0 mm may be compensated when the translational acceleration sensor is placed in such manner that one of its measurement directions is directed towards the centre of the wheel.

The wireless measurement module 408 is typically attached to the wheel rim, for instance, by means of a capsule embedded in the centre hole of the rim.

In an embodiment, the system is configured to support a variable number of measurement modules. For example, the number of measurement modules may be increased when a trailer is attached to the vehicle. The central processing unit may be configured to determine the measurement modules that are currently connected to the vehicle and to communicate with those measurement modules. This may be carried out by pairing measurement modules with the communication unit 302 of the central processing unit beforehand, and the determination may comprise detection of said paired measurement modules. The pairing may be the Bluetooth pairing or a known pairing procedure of any other wireless communication scheme. In another embodiment, the central processing unit may communicate with those measurement modules whose transmissions the communication module 302 is able to detect on a constant basis. The central processing unit may discard those measurement modules that are detected only for a (predetermined) short time period so as to exclude measurement modules of other vehicles. In connection with wired connections, the determination is simpler because of a fixed wired connection between the central processing unit 300 and the measurement modules.

Figure 4 also illustrates a coordinate system provided at the wheel 400, including a radial axis 412, a tangential axis 414 and a lateral axis, the direction of which is perpendicular to the radial axis 412 and the tangential axis 414. In addition, Figure 4 shows a coordinate system fixed in the space, including a vertical axis 420 and a horizontal axis 418. Additional directions that may be measured include the lateral direction and rotation around any one of above-mentioned axis, namely pan, tilt and roll. Acceleration of all these com- ponents may be measured with an acceleration sensor having 6 degrees of freedom, e.g. with a combination of a gyroscope and acceleration sensors.

In an embodiment, the acceleration measurement unit 202 of the wireless measurement module 408 measures radial acceleration. In this case, the acceleration information 214 includes the radial acceleration.

In an embodiment, the acceleration measurement unit 202 of the wireless measurement module 408 measures tangential acceleration. In this case, the acceleration information 214 includes the tangential acceleration.

In another embodiment, the acceleration measurement unit 202 of the wireless measurement module 408 measures lateral acceleration. In this case, the acceleration information 214 includes the lateral acceleration.

When rotating at a constant speed in the direction of the arrow 416, the wireless measurement module 408 is subjected to centrifugal acceleration in the direction of the radial axis 412 and gravitational acceleration 422 in the direction of the vertical axis 420, in which case the wireless measurement module 408 is subjected to the radial acceleration

a _m,r = ^ ² f ² r _m + g sin(4≠ + r) , (1 ) where a _{m r} is the radial acceleration measured by the wireless measurement module 408, f is the rotational frequency of the wheel 400, g=9,81 ms ^'2 is the gravitational acceleration, r _m is the measurement radius 410 and γ is the phase factor. Accordingly, the tangential acceleration may be illustrated by the expression

amj = g∞s{4≠ + y) . (2)

Figure 5 illustrates radial acceleration 502 measured by the wireless measurement module 408 as a function of time 504. The curve 506 represents the momentary radial acceleration, and the curve 508 the mean value of the radial acceleration. A similar curve may be provided for tangential acceleration.

The mean value 508 of the radial acceleration may be used for determining the first term of the right side of the equation (1), on the basis of which the rotational frequency f can be determined if the measurement radius r _m is known.

The rotational frequency may also be determined by measuring the cycle time of the curve 506 and calculating the mean value of the cycle time 510.

The rotational frequency may also be determined by measuring the cycle time of the tangential acceleration.

In an embodiment, the state determination unit 204 may register the total number of revolutions N of the wheel 400 by summing the acceleration maximums of the curve 506, for instance. In this case, the distance S that the wheel 400 has travelled may be counted on the basis of the equation

S = 2nr _w x N , (3) where r _w is the radius 406 of the wheel 400. The wheel radius 406 may be stored in the tag memory 22, for instance.

In an embodiment, the acceleration measurement unit 408 deter- mines the acceleration rotation component associated with the rotation of the wheel 400, such as radial acceleration a _{m r} or tangential acceleration a _{m t} . The acceleration measurement unit 408 supplies the value of the acceleration rotation component to the module communication unit 206, which includes the value of the acceleration rotation component in the wireless communication signal 220.

The central processing unit 300 receives the wireless communication signal 220 and supplies the value of the acceleration rotation component to the processing unit 304.

The translational acceleration sensor 306 of the central processing unit 300 may determine the acceleration translation component a _T 424 associ- ated with the translation of the vehicle wheel 102A, 102B and supply the acceleration translation component aj 424 to the processing unit 304.

The processing unit 304 may determine the value of the parameter characterizing the state or characteristic of the vehicle wheel 102A, 102B by means of the acceleration rotation component and the acceleration translation component.

The radius 406 of the wheel 400 may be determined, for instance, on the basis of the expression

where v _T is the propagation speed of the vehicle 100, which may be calculated as the time integral of the translational acceleration, for example.

With reference to the example of Figure 6, let us examine the radial acceleration of the vehicle wheel 400 in accelerated motion, measured by the wireless measurement module 408. Thus, the radial acceleration may be pre- sented by the expression

a _m,r W = 4 ² f ² (t)r _m + a _T cos(a(t))+ g sin(a(r)) (5) where (t) is the rotational angle of the wheel 400 as a function of time t.

Correspondingly, the tangential acceleration may be presented by the expression

a _{m l} (t) = a _T sin(a(t))+ g cos(a(t)) . (6)

The first term of the right side of the equation (5) is marked with a broken line 602 and it represents the centrifugal acceleration in the direction of the radial axis 412, which increases according to the translational acceleration a _T .

The second term of the right side of the equation (5) is marked with a dot-and-dash line 606 and it shows the direct effect of the translational acceleration a _T on the radial acceleration.

The third term of the right side of the equation (5) is marked with a dotted line 604 and it represents the effect of gravitation g on the radial accel- eration.

A continuous line 600 represents the radial acceleration a _{m r} of the wireless measurement module 400.

In an embodiment, the state determination unit 204 determines the acceleration translation component on the basis of the radial acceleration by adapting the curve 600 to the model according to the equation (5), for in- stance. In an embodiment, the state determination unit 204 eliminates the gravitational acceleration mathematically from the curve 600, whereby the effect of the translational acceleration remains as a periodic factor. The transla- tional acceleration may be determined on the basis of the amplitude of the curve 600 after the effect of gravitation has been eliminated. Thus, the radius 406 of the wheel 400 may be determined, for instance, by means of the equation (4) when the propagation speed v _T has been calculated as the time integral of a _T , for example. In practice, equation (4) needs not be calculated in real time but, instead, the translational acceleration may have been mapped direct- ly into the radius, rotational speed or frequency, number of revolutions of the wheel per time unit, or another parameter of the wheel according to a predetermined mapping table. The measurement module may compute the rotational frequency of the wheel, for example, from the measured radial acceleration and transmit the rotational frequency to the central processing unit. The pro- cessor of the central processing unit may then evaluate the received rotational frequency, from which the low air pressure or tread may be detected by comparing the received rotational frequency with the reference values or corresponding measurement results (measured rotational frequencies) received from one or more other wheels in order to detect a faulty wheel. The low air pressure or tread may be detected from other properties correlating with the rotational frequency. The tangential acceleration may be examined in the similar manner.

Above, the description has focused on the acceleration sensors and wheel properties that may be derived from the measured acceleration of the wheel. In other embodiments, measurement modules comprising other sensors are utilized, e.g. air pressure sensors measuring the air pressure inside each wheel, temperature sensors measuring ambient temperature and/or the temperature of brakes of each wheel so as to detect faulty brakes, and humidity sensors for detecting outdoor climate conditions and performance of the tires.

According to an aspect, the processor 304 is arranged to store previous measurement results and to predict a future event in the state of the wheel on the basis of such history data. Figure 7 illustrates a flow diagram describing a method or a computer process carrying out such functionality in an apparatus according to an embodiment of the invention. The process starts in step 700. In this embodiment, the measurement modules may be arranged to measure a plurality of measurement results related to a property of the vehicle wheel, and to transmit the measurement results to the central processing unit. The measurement module may measure any one of the above-listed properties related to the wheels, e.g. pressure, acceleration, temperature, and humid- ity. The measurement results may be actual measured values, i.e. raw measurement data, or they may comprise a parameter that has been computed from the measured raw data. The processor 304 of the central processing unit 300 is arranged to receive the measurement results from the at least one measurement module (104A, 104B, 200, 408) in step 702. When the measurement results are raw data, the processor may process the raw measurement data so as to derive any one of the above-described properties derivable from the measurements, e.g. tyre pressure from raw acceleration measurement data.

In step 704, the processor constructs a temporal model on the basis of the received measurement results. The temporal model may comprise said received measurement results as a function of measurement time. The measurement time may be provided by the measurement module(s), in which case the measurement time reflects the time when the measurement is carried out in the measurement module. In another embodiment, the measurement time may be provided by the processor in which case the measurement time re- fleets the time when the measurement result is received by the processor. The measured properties, e.g. tyre pressure, typically change over a long period of time (days, weeks, months, etc.). The processor 304 is configured to construct the temporal model for each wheel and for each measured property, and to predict on the basis of the constructed temporal model a future event related to the measured property of the vehicle wheel in block 706. Then, the process returns to step 702 where new measurement results are received, and the temporal model is updated by taking into account the newly received measurement values. The updating may utilize all or some of the previous measurement results, and the new results may be used to improve the accuracy of the temporal model. In an embodiment, older measurement results are discarded as new measurement results are obtained so as to provide a sliding temporal model. The processor may also be provided with a reference model (stored in the memory unit) based on the wheel type, which will improve the predictability of the temporal model.

Figure 8 illustrates exemplary graphs for such temporal models for four wheels. The received measurement results are denoted by "x". The measurement timings of the measurement modules may be synchronized, or they may report the measurement results independently. The central processing unit may request the measurement modules to carry out the measurement and report the measurement results, or the measurement modules may determine the measurement timing and reporting timing independently, e.g. periodical measurements. The processor 304 may be configured to compute the temporal model from the received measurement results by computing a least squares (LS) solution or another equivalent (linear or non-linear) regression model that enables construction of a mathematical function describing a correlation and a tendency that is derivable from the measurement results. Graphs 800 to 806 in Figure 8 illustrate such a regression model for each wheel. The horizontal axis shows the time, and the vertical axis shows the measured property, e.g. tyre pressure, wheel radius, or any property derived from the acceleration information as described above. Three graphs 802, 804, 806 show substantially similar temporal models, while one of the graphs 800 exhibits a tendency indicating a faulty wheel. At present time, the difference in the measured property may still be within tolerable region, but from the predicted future behaviour of the graph 800 it may be foreseen that the wheel may become faulty, e.g. lose pressure or break, and possibly cause an accident. As the temporal model enables prediction (or of the future behaviour of the measured property by extrapolating a tendency derived from the measurement results into the future, the processor is able to detect occurrence of such an event beforehand and alert the operator in time before the fault realizes.

In practice, the processor may be configured to compare the tem- poral models associated with said different wheels, and to predict the future event on the basis of the comparison of the temporal models. The processor may be configured to compare the difference in the measured property between the wheels, and to predict, on the basis of said temporal models, a time instant TO when the difference in the measured property of at least one wheel with respect to the measured property of one or a plurality of other wheels will exceed a predetermined threshold. The comparison of the temporal model of a given wheel may be made with respect to a reference wheel, the other wheels individually, and/or an average of the other wheels. In an embodiment, the occurrence of the upcoming event may be detected on the basis of comparison with the reference wheel and optionally one other wheel so as to verify the event, i.e. to verify which one of the wheels is the faulty one as described above, and the time instant TO for the event may be computed by comparing the temporal model of the future faulty wheel with an average temporal model of the other wheels where the average temporal model is computed from the temporal models of the other wheels by averaging them into a single average temporal model. In all embodiments, the processor may change the reference wheel, or it may use multiple wheels as reference wheels simultaneously.

In response to determining the time instant TO, the processor may be configured to make a report through the user interface, for example, so as to notify that the operator should check the issue before the time instant TO. The processor may be configured to carry out the report at a time derived from the time instant TO, e.g. two weeks before the time instant TO, and the report may indicate that the issue should be checked within a week. In general, the report may indicate a time interval in which the operator should check the issue. The processor may also be configured to store and/or report when the prediction has been executed previously. The prediction and/or reporting may be triggered on the basis of time and/or travel distance of the vehicle which ever expires first, e.g. weekly and/or after every thousand kilometres.

In some embodiments, external input data may be used when predicting the future events on the basis of the measured properties. Such input data may include location of the vehicle in terms of absolute location and/or distance to a destination. The location may be determined by using a positioning system known in the art, e.g. a global positioning system (GPS). Then, the processor may check whether the predicted event occurs before an estimated time of arrival at the destination. If the predicted event occurs before an esti- mated time of arrival at the destination, the processor may trigger a ^' first type of alarm requesting immediate measures from the operator. If the predicted event occurs after the estimated time of arrival at the destination, the processor may trigger another type of alarm which informs the operator that the vehicle needs maintenance upon arriving at the destination. In another embodi- ment, the input data includes weather/road conditions that may be received through a public or private radio telecommunication system (GSM, UMTS, etc.). In poor weather/road conditions, the processor may advance the time instant when the alarm for the predicted event is triggered. In good weather/road conditions, the processor may maintain or even delay the time instant when the alarm for the predicted event is triggered.

Such external input data related to the vehicle (e.g. speed) and/or environment (e.g. location, weather conditions) can be used to further improve the alarm and prediction of the wheel problems. For example, if road conditions are poor, the threshold for triggering the alarm will be lower and if it is known that the road is in poor shape, it can be informed to the driver. In a fur- ther example the predicted time instant for wheel check is sooner, if speed of the vehicle is higher and/or the weather/road conditions are poor. In other words, the processor may prompt the measurement modules to carry out the measurements sooner when the speed is high and/or the weather/road conditions are poor and vice versa. This check can be periodic with adjustable peri- odicity and, thus, the system may change the prediction rate (periodicity) depending on the speed and/or the current weather/road conditions so as to adapt the reliability of the prediction to the current environment. Accordingly, the system improves the safety and prediction reliability in poor conditions by using additional statistics, i.e. more data.

In summary, the present invention may be used to predict future events that may be hazardous for the vehicle and the driver beforehand so that such events may be avoided. As a consequence, the present invention alerts the operator beforehand upon detection of such an event in the future and, thus, the present invention improves safety of the vehicles and may even save lives. An embodiment where the processor compares the measurement results obtained from different wheels with each other has the advantage that ambient properties such as the temperature and humidity do not interfere the comparison, as the ambient properties are the same for each wheel. Therefore, an unreliable temperature compensation etc. to calibrate the measurement results is not necessary, and the procedure is simplified and the accuracy is improved. Additionally, the inherent properties of the wheels or tyres, e.g. wheel size or type, do not affect the reliability of the measurements, and the processor needs not preliminary information on the tyre to carry out the evaluation. The present invention works even if the vehicle is equipped with wheels of different sizes at the same time, because it does not need to compare the absolute values from the wheels, but changes in them.

Although the invention was described above with reference to the example according to the accompanying drawings, it is clear that the invention is not restricted thereto but may be modified in various ways within the scope of the appended claims.