TIRE-SIDE DEVICE AND TIRE TOOL CONTAINING THIS

Abstract

A tire-side device (1) attached to a tire (3) of a vehicle and used to measure the condition of a road surface on which the vehicle travels, wherein the tire-side device (1) comprises: a vibration detection device (10) configured to output a detection signal according to a vibration level of the tire (3); a control device (11) comprising a feature set extraction device (11a) configured to extract a feature set from the detection signal upon rotation of the tire (3); and a transmission device (12) configured to transmit road surface data containing the feature set extracted by the feature set extraction device (11a), wherein the feature set extracted by the feature set extraction device (11a) upon a previous rotation of the tire (3)a past feature set, wherein the feature set extracted by the feature set extraction device (11a) during a given rotation of the tire (3) is a present feature set, wherein the control device (11) further comprises: a feature set memory (11b) configured to store the past feature set, a variation determination device (11c) configured to determine, based on the past feature set and the present feature set, that a variation occurs in the state of the road surface, and a transmission control device (11d) configured to control a transmission of the road surface data and to cause transmission device (12) to transmit the road surface data containing the present feature set in response to the variation determination device (11c) determining that the change occurs in the state of the road surface.

Description

CROSS REFERENCE TO RELATED APPLICATIONThis application is based on Japanese Patent Application JP 2017 - 087 554 filed on Apr. 26, 2017, the disclosure of which is incorporated herein by reference.TECHNICAL FIELDThe present disclosure relates to a tire-side apparatus and a tire apparatus. The tire-side device detects vibrations of a tire and generates road surface data indicating a road condition based on the vibration data, and then transmits the road surface data to a body-side system. The tire apparatus includes a road surface state estimation function for estimating a road surface state based on the road surface data.PRIOR ARTUS 2016 / 0 368 502 A1 discloses a road surface state estimation device that estimates the state of a road surface based on vibrations acting on the tires of a vehicle without performing frequency analysis, so that the number of frequency analysis components and the power consumption are reduced.In WO 2016 / 092 824 A1, a vehicle control device is shown which controls the acceleration and deceleration of a vehicle in order to avoid a collision with a preceding vehicle or the like, wherein the acceleration and deceleration of the vehicle can be better controlled based on changes in the coefficient of friction μ between a tire and a road surface.Conventionally, JP 2016-107 833 A describes that a rear surface of a tire tread includes an acceleration sensor, and the acceleration sensor detects vibrations applied to a tire. Patent Literature 1 describes a method of determining the state of the road surface to estimate a state of the road surface based on a detection result of the vibration. In the road surface state determination method, the road surface state is determined by extracting a feature vector from the vibration waveform of the tire detected by the acceleration sensor and calculating the degree of similarity between the extracted feature vector and all the stored carrier vectors for classification of the road surface. For example, the degree of similarity between the extracted feature vector and all the carrier vectors is calculated using a kernel function, and the type of road surface such as a dry road surface or a wet road surface having the highest degree of similarity is determined as the present road surface. It may be possible to execute such a method for determining the state of the road surface for determining the road surface with high robustness.SUMMARY OF THE INVENTIONWith respect to the method for determining the state of the road surface described in JP 2016-107 833 A, it is not certain whether the calculation of the degree of similarity is performed by a controller in a tire or in a body-side system. Since the load in calculating the degree of similarity is high, the following situation may occur.For example, the in-tire controller calculates the degree of similarity and determines the state of the road surface. In a situation where a determination result is transmitted to the body-side system, it is necessary to store the carrier vector in the controller included in the tire and execute the data processing for calculating the degree of similarity. As a result, in the tire, power consumption increases greatly, and the amount of storage consumption for enormous data storage and data processing also increases greatly.In addition, the in-tire controller extracts only the feature vectors and transmits the data to the body-side system. The body side system calculates the degree of similarity against all the carrier vectors to determine the condition of the road surface. In this situation, since the timing of data transmission cannot be determined, the communication frequency must be increased. Therefore, the high-frequency data transmission increases the value of power consumption.An object of the present disclosure is to provide a tire-side device for saving storage and performance of a controller in a tire, and a tire apparatus including the tire-side device.This object is achieved by the tire-side device according to claim 1 and the tire apparatus according to claim 5. Developments according to the invention are the subject matter of the dependent claims.A tire-side apparatus according to an aspect of the present disclosure includes: vibration detection means that outputs a detection signal according to a level of vibration of a tire; control means that includes feature amount extraction means for extracting a feature amount of the detection signal in one rotation of the tire; and transmission means that transmits road surface data including the feature amount extracted by the feature amount extraction means. The feature amount extracted by the feature amount extraction device at a past rotation of the tire is a past feature amount, and the feature amount extracted by the feature amount extraction device at a present rotation of the tire is a present feature amount. The controller also includes: a feature amount memory configured to store past feature amounts; a variation determiner configured to determine a variation in the state of the road surface based on the past feature amount and the present feature amount; and a transmission controller configured to control the transmission of the road surface data and cause the transmitter to transmit the road surface data including the present feature amount in response to the variation determiner determining that the variation occurs in the state of the road surface.The variation determination device determines variation of the state of the road surface based on a present feature amount and a past feature amount, and transmits the road surface data having the present feature amount in a situation where the variation determination device determines that there is variation of the state of the road surface. Therefore, it is possible to lower the communication frequency and achieve saving in power consumption in the control device included in the tire. Since it is not necessary for the controller in the tire-side apparatus to include a carrier vector memory for storing the carrier vectors, it is also possible to save storage space in the controller included in the tire. Reference numerals with parentheses attached to the respective constituent elements and the like show an example of a correspondence relationship between the constituent elements and the like and specific constituent elements and the like described in the embodiment described later.BRIEF DESCRIPTION OF THE DRAWINGSFIG. 1 is a view showing a block configuration of a tire apparatus to which a tire-side device according to a first embodiment is applied in a vehicle-mounted state. FIG. 2 is a block diagram showing the details of the tire-side device and a body-side system. FIG. 3 is a schematic cross-sectional view of a tire to which a tire-side device is attached. FIG. 4 is a diagram showing an output voltage waveform of an acceleration detection device during tire rotation. FIG. 5 is a diagram illustrating a detection signal of the acceleration detection device divided by each of the time slots having a predetermined time width T. FIG. 6 shows a relationship between determinants Xi(r), Xi(r-1) and a distance K yz in each area obtained by dividing a time axis waveform at a time of present rotation of the tire and a time axis waveform at a time of rotation before present rotation of the tire with each of the time windows having the predetermined time width T. FIG. 7 is a flowchart showing a data transfer process executed by a controller in the tire-side device. FIG. 8 is a flowchart showing a process for estimating the state of the road surface, which is executed by a controller in a body-side system.EMBODIMENTS FOR CARRYING OUT THE INVENTIONEmbodiments of the present disclosure will be described below with reference to the drawings. In the following respective embodiments, parts that are identical or equivalent to each other are denoted by the same symbols for description.(First Embodiment)The following describes a tire apparatus 100 including a road surface state estimation function according to the present embodiment with reference to FIGS. 1 to 8 The tire apparatus 100 according to the present embodiment estimates a road surface state during traveling of a vehicle based on vibrations applied to a ground-contacting surface of a tire included in each of the wheels of the vehicle.The tire apparatus 100 shown in FIGS. 1 and 2 includes a tire-side device 1 disposed on a wheel and a body-side body system 2 including a plurality of devices included in the vehicle. The body-side system 2 includes, for example, a receiver 21, a brake control electronic control device 22 (hereinafter, referred to as a "brake ECU"), and a notification device 23.The tire apparatus 100 according to the present embodiment transmits, from the tire side device 1, data according to a state of the road surface (hereinafter, referred to as "road surface data") during which the tire 3 travels. The tire apparatus 100 receives the road surface data through the receiving means 21 to estimate the state of the road surface. The tire apparatus 100 transmits an estimation result of the state of the road surface at the receiving device 21 to the notifying device 23, and then notifies the estimation result of the state of the road surface through the notifying device 23. It is also possible to warn the driver when the roadway is slippery. The tire apparatus 100 transmits a state of the road surface to the brake ECU 22 or the like for executing vehicle movement control, so that the vehicle movement control can be executed for avoiding an undesired situation. For example, in the case of a frozen road, a braking force generated in relation to the amount of braking is weakened compared to a situation of the dry road surface. Thus, the vehicle movement control corresponding to a situation where the road surface μ is lower can be achieved. The tire-side apparatus 1 and the receiver 21 are configured as follows.As illustrated in FIG. 2, the tire-side device 1 includes an acceleration detection device 10, a controller 11, and a data communication device 12.The acceleration acquisition device 10 includes vibration detection means for detecting vibrations applied to the tire 3. For example, the acceleration acquisition device 10 is configured by an acceleration sensor. In a situation where the acceleration acquisition device 10 is the acceleration sensor, the acceleration acquisition device 10 outputs a detection signal of the acceleration as a detection signal according to the vibration in a direction in contact with a circular trajectory drawn from the tire side device 1 during the rotation of the tire 3. In other words, the acceleration acquisition device 10 outputs the detection signal of the acceleration as a detection signal corresponding to the vibration in a direction of the tire tangential direction, as indicated by an arrow X in FIG. 3.The controller 11 corresponds to a first controller and is configured by a microcomputer having, for example, a CPU, a ROM, a RAM, an I / O. The controller 11 is an apparatus for executing the above processes according to a program stored in the ROM or the like. The controller 11 includes a feature amount extraction device 11 a, a feature amount storage 11 b, a variation determination device 11 c, and a transmission controller 11 dconfigured as function devices for executing the above processes.The feature amount extraction device 11 auses the detection signal output from the acceleration acquisition device 10 as the detection signal indicating vibration data in the tire tangential direction to process the detection signal for extracting the feature amount of the tire vibration. In the present embodiment, the detection signal of the acceleration of the tire 3 (hereinafter, referred to as "tire G") is processed by signal processing to extract the feature amount of the tire G. The feature amount extraction device 11 atransmits data having the extracted feature amount as road surface data to the data communication device 12 through the transmission controller 11 d. The details of the feature amount will be described.The feature amount storage 11 bstores a feature amount extracted by the feature amount extraction device 11 abefore the tire 3 makes a rotation (hereinafter referred to as "previous feature amount"). Since rotation of the tire 3 can be confirmed by a method described below, the feature amount is stored in a rotation unit every time the tire 3 makes rotation. The feature amount in a rotation unit of the tire 3 can be updated every time the tire 3 makes a rotation. Alternatively, a plurality of rotation units are stored, and the oldest data may be deleted every time the tire 3 makes one rotation. In consideration of saving storage space for the controller 11 in the tire 3, it is preferable to decrease the amount of stored data. Therefore, it is preferable to update the data each time the tire 3 makes a rotation.The variation determiner 11 cdetermines whether there is variation in the state of the road surface based on the feature amount extracted by the feature amount extractor 11 ain the present rotation of the tire 3 (hereinafter, also referred to as "present feature amount") and the previous feature amount of the tire 3 stored in the feature amount storage 11 b. When the variation determiner 11 cdetermines that there is variation in the state of the road surface, the variation determiner 11 ctransmits, to the transmission controller 11 d, a control signal indicating the variation.The transmission controller 11 dcontrols data transmission from the data communication device 12. the transmission controller 11 d receives, from the variation determiner 11 c, the control signal indicating variation of the state of the road surface, and then the transmission controller 11 dtransmits, to the data communication device 12, road data including the present feature amount extracted at this time by the feature amount extractor 11 a.The data communication apparatus 12 includes a transmission means. For example, when the data communication device 12 receives the road surface data from the transmission controller 11 d, the data communication device 12 performs the transmission of the road surface data including the present feature amount at this timing. Via the timing of data transmission from the data communication device 12, the timing is controlled by the transmission controller 11 d. Thus, the data transmission is not performed every time the tire 3 makes a rotation, but the data transmission is performed only when the state of the road surface varies.On the other hand, the receiving means 21 includes a data communication apparatus 24 and a control means 25 as shown in FIG. 2.The data communication device 24 includes a receiving device. The data communication device 24 receives the road surface data including the present feature amount transmitted from the data communication device 12 in the tire-side device 1, and transmits the road surface data to the controller 25.The control device 25 corresponds to a second control device. The controller 25 is provided by a microcomputer including a CPU, a ROM, a RAM, an I / O, or the like, and executes a plurality of processes for detecting the state of the road surface based on a program stored in the ROM or the like. The controller 25 includes a carrier vector memory 25 aand a state estimation device 25 bas functional devices for executing a plurality of processes.The carrier vector memory 25a stores a carrier vector for classifying the type of the road surface. The carrier vector is a model feature amount obtained by learning using, for example, a carrier vector machine. A vehicle having the tire-side device 1 is experimentally operated on various kinds of road surfaces, and the feature amount extracted by the feature extraction device 11 ain a predetermined number of times of tire rotation is learned. The carrier vector is a part obtained by extracting a typical feature amount from the learned amount in a predetermined number of parts. For example, the feature amount in one million units of rotation is learned in each type of road surface, and the part obtained by extracting a typical feature amount in one million units of rotation from the learned amount is configured as a carrier vector.The state estimation device 25 bcompares the present feature amount received from the tire side device 1 by the data communication device 24 with a carrier vector for classifying the type of the road surface stored in the carrier vector memory 25 a, and estimates the state of the road surface. For example, the present feature amount is compared with the carrier vector for classification of the road surface type, and the road surface corresponding to the carrier vector having the highest similarity with the present feature amount is estimated as the present road surface.When the controller 25 estimates the state of the road surface, the controller 25 transmits the estimated state of the road surface to the notifier 23, and notifies a driver of the state of the road surface, if necessary, through the notifier 23. For example, the estimated state of the road surface may always be displayed by the notifying device 23, or the road state may be displayed to notify the driver only when the vehicle needs to be driven more gently than usual, for example, when the estimated road state corresponds to the wet road or the icy road. Moreover, when the transceiver 21 transmits the state of the road surface to an ECU for vehicle motion control such as the brake ECU 22, the vehicle motion control may be executed based on the transmitted state of the road surface.The brake ECU 22 includes a brake control device that performs various brake controls. Specifically, the brake ECU 22 increases or decreases a wheel cylinder pressure by driving a brake fluid pressure control actuator to control a braking force. Moreover, the brake ECU 22 may control the braking force of each wheel independently of each other. When the state of the road surface is transmitted from the receiver 21 via the brake ECU 22, the control of the braking force is executed as motion control of the vehicle based on the state of the road surface. For example, when the brake ECU 22 indicates that the transmitted road surface is a frozen road, the brake ECU 22 weakens the generated braking force with respect to the braking action performed by the driver as compared to a situation on the dry road surface. This makes it possible to prevent wheel slip and to avoid possible undesired situations for the vehicle.The notifying device 23 is configured with, for example, a meter display device, and is used to notify the driver of the state of the road surface. In the case that the notification device 23 is configured with the meter display device, it is located at a position that the driver can recognize during the travel of the vehicle, for example, inside an instrument panel in the vehicle. When the state of the road surface is transmitted from the transceiver 21, the counter display device visually notifies the driver of the state of the road surface by performing display such that the state of the road surface can be interpreted.The notification device 23 may also be configured with, for example, a buzzer or a voice guidance device. In this case, the notifying device 23 audibly notifies the driver of the state of the road surface through buzzer sound or voice guidance. Although the meter display device has been exemplified as the visual notification device 23, the notification device 23 may be configured with a display device that displays information such as a head-up display.The tire apparatus 100 according to the present embodiment is configured as described above. Each device constituting the body-side system 2 is connected via an in-vehicle LAN (Local Area Network) such as CAN (Controller Area Network). Therefore, each device can communicate information with each other via the body-side LAN.The following describes the details of the feature amount extracted by the feature amount extraction device 11 aand the determination of variation in the state of the road surface by the variation determination device 11 c.First, the feature amount extracted by the feature amount extraction device 11 awill be described. The feature amount is an amount indicating the feature of the vibration applied to the tire 3. The feature amount is obtained by the acceleration detection device 10.The output voltage waveform of a detection signal in the acceleration detection device 10 during the tire rotation is, for example, a waveform shown in FIG. 4. As illustrated in the drawing, the output voltage of the acceleration detecting device 10 has a maximum value at the contact start time when the portion of the tread 31 of the tire 3 corresponding to the position of the acceleration detecting device 10 starts to contact the ground with the rotation of the tire 3. Hereinafter, a peak value at the start of the ground touch at which the output voltage of the acceleration detection device 10 has the maximum value is referred to as a first peak value. As shown in FIG. 4, the output voltage of the acceleration detecting device 10 has a minimum value at the contact end time when the portion of the tread 31 of the tire 3 corresponding to the position of the acceleration detecting device 10 is changed from a state in which the ground is touched to a state in which the ground is not touched with the rotation of the tire 3. Hereinafter, this peak value at the end of the ground touch at which the output voltage of the acceleration detection device 10 has the maximum value will be referred to as a second peak value.The reason why the output voltage of the acceleration detection device 10 has a peak value at the timing is as follows. When the portion of the tread 31 corresponding to the position of the acceleration detecting device 10 contacts the ground with the rotation of the tire 3, a portion of the tire 3 having a substantially cylindrical surface is pressed to be deformed into a flat shape in the vicinity of the acceleration detecting device 10. After receiving the surge at this time, the output voltage of the acceleration detection device 10 has the first peak value. When the portion of the tread 31 corresponding to the position of the acceleration detector 10 is separated from the ground with the rotation of the tire 3, the pressing of the portion of the tire 3 in the vicinity of the acceleration detector 10 is released, and the flat shape of the portion of the tire 3 returns to the substantially cylindrical shape. After receiving the impact, when the shape of the tire 3 returns, the output voltage of the acceleration detection device 10 has the second peak value. The output voltage of the acceleration detection device 10 takes the first peak value and the second peak value at the ground touch start time and the ground touch end time, respectively. Since a direction of impact when the tire 3 is pressed and a direction of impact when the tire 3 is released from the pressurization are opposite, the polarities of the output voltages are also opposite.The portion of the tire tread 31 corresponding to a location where the acceleration detection device 10 is disposed at a moment of ground contact is referred to as a stepping zone. The portion of the tire tread 31 at a moment of leaving the road surface is referred to as a "kick-out zone". The "step zone" includes a timing with the first peak value. The "detachment zone" includes a timing with the second peak value. The region before the step region is referred to as a "pre-stepping region". The zone from the step zone to the detachment zone, i.e., the portion of the tire tread 31 corresponding to the attachment position of the acceleration detection device 10 during ground contact is located in a zone referred to as a "pre-kick out zone" and a "pre-detachment zone", respectively. The zone after the detachment zone is referred to as "post-kick out zone" or "after-detachment zone". A period during which the portion of the tire tread 31 corresponding to the attachment position of the acceleration detection device 10 is in contact with the ground, and both ends of the period can be divided into five zones in total. FIG. 4 illustrates that "pre-stage zone", "stage zone", "pre-detachment zone", "detachment zone", and "post-detachment zone" of the detection signal are displayed as five zones R 1 to R 5, respectively, in the order.The vibration generated in the tire 3 in each divided zone varies, and the detection signal of the acceleration detection device 10 varies depending on the state of the road surface. When frequency analysis of the detection signal of the acceleration detection device is performed in each zone, the state of the road surface on which the vehicle travels can be detected. For example, a shear force at the time of peeling decreases in a situation of slippery road surfaces such as a snowy road. Therefore, a bandwidth value selected from a 1 kHz to 4 kHz band decreases in the peeling zone R 4 and the post-peeling zone R 5. Since a plurality of frequency components of the detection signal in the acceleration detection device 10 vary depending on the state of the road surface, it is possible to determine the state of the road surface based on the frequency analysis of the detection signal.The feature amount extraction device 11 adividifies the detection signal of the acceleration detection device 10 having a continuous-time waveform for one rotation of the tire 3 into a plurality of sections. Each section has a time window with a predetermined time width T as shown in FIG. 5. The feature amount extraction device 11 aexecutes the frequency analysis in each region to extract the feature amount. Specifically, when the frequency analysis is performed in each region, a power spectral density level in each frequency band, i.e., an oscillation level in a specific frequency band, is evaluated. The power spectral density level represents the feature amount.The number of regions divided by the time window having a time width T is variable according to a vehicle speed, particularly a rotational speed of the tire 3. In the following description, "n" represents the number of regions for one rotation. "n" is a natural number.For example, detection signals in the respective regions pass through corresponding multiple filters having specified frequency bands to obtain power spectral density levels. The plurality of filters may be, for example, five band pass filters having respective frequency bands of 0 to 1 kHz, 1 kHz to 2 kHz, 2 kHz to 3 kHz, 3 kHz to 4 kHz, and 4 kHz to 5 kHz. The power spectral density levels represent the feature amount. The feature set is called a feature vector. When the power spectral density level in each specified frequency band is indicated as a ik the feature vector Xirepresents, in a region "i" in which "i" is a natural number and is in a relationship of 1≤i≤n, the following expression as a determinant with the power spectral density levels as elements. "k" in the power spectral density level a ik represents the number of the specified frequency bands, in other words, the number of the band pass filters. In a situation where a 0 to 5 kHz band is divided into five sections, "k" is in a ratio of k=1 to 5.The determinant X is an expression representing the feature amount for rotation of the tire. The feature amount extraction device 11 aextracts the feature amount expressed in the determinant X by performing the frequency analysis of the detection signal of the acceleration detection device 10.Next, the determination of variation of the state of the road surface by the variation determination device 11 cis described. The determination is performed by calculating a degree of similarity based on the present feature amount extracted by the feature amount extraction device 11 aand the previous feature amount stored in the feature amount storage 11 b.With respect to the determinant X representing the above-described feature amount, X(r) represents the determinant of the present feature amount, and X(r-1) represents the determinant of the previous feature amount. The respective expressions a(r) ik, a(r-1) ik represent the power spectral density level a ik as an element of each determinant. In this situation, the determinant X(r) of the present feature amount and the determinant X(r-1) of the previous feature amount are respectively expressed as follows.The degree of similarity indicates a degree of similarity between the feature sets represented by two determinants. The two determinants are similar as the degree of similarity is higher. In the present embodiment, the variation determination device 11 cvaluates the degree of similarity using the kernel method and determines the variation of the state of the road surface based on the degree of similarity. An inner product of the determinant X(r) at the time of the present rotation of the tire 3 and the determinant X(r-1) at the time of rotation before the present rotation is calculated, and the inner product is used as the degree of similarity. In other words, the distance between the coordinates indicated by the feature vector Xi in the sections divided by each time window having a predetermined time width T within a feature space is calculated, and the distance is used as the degree of similarity.For example, as shown in FIG. 6, with respect to the time axis waveform of the detection signal in the acceleration detection device 10, the time axis waveform at the present rotation of the tire 3, and the time axis waveform at a rotation before the present rotation are each divided into ranges by the time window having a predetermined time width T. As shown in the drawing, since each of the time axis waveforms is divided into five regions, "n" is in a ratio of n=5, and "i" is expressed by a ratio of 1≤i≤5. As shown in the drawing, Xi(r) represents the feature vector in each region at the present rotation, and Xi(r-1) represents the feature vector in each region at the time of rotation before the present rotation. In this situation, with respect to the distance K yz between the coordinates indicated by the feature vectors Xi in the regions, it is indicated that a horizontal tile including the feature vector Xi(r) in each region of the present rotation is intersected with a vertical tile including the feature vector Xi(r-1) in each region at the time of rotation before the present rotation. With respect to the distance K yz "y" is a value obtained by rewriting "i" to Xi(r-1), and "z" is a value obtained by rewriting "i" to Xi(r). Since the vehicle speed does not vary greatly between the time of the present rotation and the time of a rotation before the present rotation, the number of ranges at each rotation is basically similar.In the present embodiment, the feature vector is obtained by dividing into five specified frequency bands. The feature vector Xi in each region is expressed in a six-dimensional space combined with the time axis. The distance between the coordinates indicated by the feature vectors Xi in the respective regions is the distance between the coordinates in the six-dimensional space. The distance between the coordinates indicated by the feature vectors in the respective sections decreases when the feature amounts are similar, and increases when the feature amounts are not similar. Thus, the degree of similarity increases as the distance decreases, and the degree of similarity decreases as the distance increases.For example, in a situation where the regions 1 to n are obtained by time division, the distance K yz between the coordinates indicated by the feature vector in the region 1 is expressed as follows.The distance K yz between the coordinates indicated by the feature vectors in the respective regions is evaluated for all the regions, and then the sum of the distance K yz is calculated for all the regions. The sum K total is used as a value corresponding to the degree of similarity. The sum K total is compared with a predetermined threshold value Th. When the sum K total is larger than the threshold Th, the degree of similarity is smaller and it is determined that there is variation in the state of the road surface. When the sum K total is smaller than the threshold Th, the degree of similarity is smaller and it is determined that there is no variation in the state of the road surface.The sum K total of the distances K yz between two coordinates indicated by the feature vectors in the respective regions is used as a value corresponding to the degree of similarity. Other values may also be used as parameters for indicating the degree of similarity. For example, the average distance K ave, which is an average value of the distance K yz, which is evaluated by dividing the sum K total by the number of regions, may be used as a parameter indicating the degree of similarity. As shown in Patent Literature 1, the degree of similarity can also be evaluated with different kernel functions, and the degree of similarity can be calculated except for a path having a low degree of similarity.Next, an operation of the tire apparatus 100 according to the present embodiment will be described with reference to FIG. 7.The controller 11 in the tire-side apparatus 1 of each wheel executes a data transfer process shown in FIG. 7. The process is executed for each predetermined control cycle.In step S 100, input processing for the detection signal is performed in the acceleration detection device 10. The processing is continued for a period until the tire 3 makes a rotation in a subsequent step S 110. When the detection signal is input to the tire rotation acceleration detector 10, the process proceeds to a subsequent step S 120, and then the feature amount of the time axis waveform of the detection signal is extracted in the tire rotation acceleration detector 10. The processing from steps S 100 to S 120 is performed by the feature amount extraction device 11 a.Rotation of the tire 3 is determined based on the time axis waveform of the detection signal in the acceleration detection device 10. Since the detection signal represents the time axis waveform shown in FIG. 4, rotation of the tire 3 can be interpreted by checking the first peak value and the second peak value of the detection signal.The state of the road surface appears particularly as a variation of the time axis waveform of the detection signal between both ends of the period including "step zone", "pre-detachment zone", and "detachment zone". Therefore, the data can be input during this period. It is not necessary to input all data of the detection signal to the acceleration detection device 10 during rotation of the tire. For example, in the "pre-stage zone" and "after-detachment zone", the data may be input in the vicinity of the "pre-stage zone" and "after-detachment zone" when the data is present. Therefore, in a region where the vibration level of the detection signal in the acceleration detection device 10 is lower than a threshold value, the detection signal may not be input during the "pre-stage region" and "post-detachment region", because these periods are hardly affected by the influence of the state of the road surface.The extraction of the feature amount at step S 120 is performed according to the above-described method.Subsequently, the process proceeds to step S 130. The degree of similarity is evaluated according to the above-described method based on the present feature amount and the previous feature amount, and for example, the degree of similarity is compared with the threshold value Th to determine whether there is a variation in the state of the road surface. Since this processing is performed by the variation determination device 11 c, this processing is executed based on the present feature amount extracted by the feature amount extraction device 11 aand the previous feature amount stored in the feature amount storage 11 bat step S 150.In a situation where an affirmative determination is made in step S 130, the road surface data including the present feature amount is transmitted from the transmission controller 11 dto the data communication device 12 for executing the data transmission in step S 140. Therefore, the road surface data including the present feature amount is transmitted by the data communication device 12. The road surface data including the present feature amount is transmitted from the data communication device 12 only when there is a change in the state of the road surface. The data transfer is not carried out when there is no change in the state of the road surface. Thus, it is possible to reduce the communication frequency, and it is possible to achieve saving in power consumption of the controller 11 in the tire 3.Finally, the process proceeds to step S 150. The present feature amount is stored in the feature amount storage 11 bas a previous feature amount, and the process ends.On the other hand, the controller 25 of the receiver 21 executes a road surface estimation process shown in FIG. 8. The process is executed for each predetermined control cycle.First, in step S200, the data reception processing is executed. The processing is performed by the controller 25 which fetches the road surface data when the data communication device 24 receives the road surface. When the data communication device 24 does not perform data reception, the controller 25 ends the processing without retrieving road surface data.Subsequently, the process proceeds to step S 210 and determines whether there is data reception. In a situation where there is data reception, the process proceeds to step S 220. In a situation where there is no data reception, the processing in steps S 200, S 210 is repeated until the data reception occurs.The process proceeds to step S220, and estimation of the state of the road surface is performed. With respect to the estimation of the state of the road surface, the state of the road surface is estimated by comparing the present feature amount included in the received road surface data with a carrier vector for classifying the road surface stored in the carrier vector memory 25a. For example, the degree of similarity between the present feature amount and all the carrier vectors is evaluated to classify the road surface. The road surface corresponding to the carrier vector having the highest similarity is estimated as the present road surface. For the calculation of the degree of similarity in this situation, the similar method of calculating the present feature amount and the previous feature amount may be applied in step S 130 of FIG. 7.As described above, it is possible to estimate the state of the road surface on which the vehicle travels with the tire apparatus 100 according to the present embodiment. In estimating such a state of the road surface, the transmission of road surface data including the present feature amount from the tire-side device 1 is executed only at a time when there is a change in the state of the road surface. Therefore, it is possible to reduce the communication frequency, and it is possible to achieve saving in power consumption of the controller 11 in the tire 3.Since the tire apparatus 100 does not require to include the carrier vector memory for storing the carrier vector in the controller 11 of the tire-side apparatus 1, it is possible to store the memory of the controller 11 in the tire 3.With respect to the data processing for calculating the degree of similarity by the controller 11 of the tire-side device 1, the data processing may be applied only to the present feature amount and the previous feature amount. With respect to calculating the degree of similarity between the present feature amount and all the carrier vectors, it may be executed on the body-side system 2. Therefore, it is possible to further suppress the amount consumed in the memory of the controller 11 in the tire 3, so that saving of the memory consumption can be achieved.It is possible to provide the tire-side device 1 in the tire 3 for saving memory and power consumption of the controller 11 and the tire apparatus 100 including the tire-side device 1.(Other Embodiments)Although the present disclosure has been described in accordance with the above-described embodiments, the present disclosure is not limited to the above-described embodiments and includes various modifications and variations within the scope of equivalents. In addition, various combinations and modes and other combinations and modes including only a single element or more or less elements are all included in the scope and breadth of the spirit of the present disclosure.For example, the feature amount storage 11 bstores the feature amount one rotation before the present rotation as the feature amount at the time of past rotation of the tire 3, but it is not necessary to have only the feature amount at one rotation before the present rotation. The feature amount storage 11 bis not limited to storing the previous feature amount as the feature amount in the past rotation of the tire 3 (hereinafter, referred to as "past feature amount"). A plurality of feature amounts in the respective multiple rotations before the present rotation may be stored as a past feature amount, and an average value of the plurality of feature amounts including the previous feature amount may also be stored as a past feature amount. For calculation of the degree of similarity with the previous feature amount, the previous feature amount of the past feature amounts may be used, and the average value of the past feature amounts including the previous feature amount may also be used. In view of saving storage space, it may be reasonable to store a minimum number of feature amounts.Although the above embodiment shows that the vibration detecting means is configured by the acceleration detecting apparatus 10, the vibration detecting means may be configured by any other vibration detecting means, e.g., a piezoelectric element.When there is variation in the state of the road surface, the road surface data including the present feature amount is transmitted from the tire-side device 1. However, the previous feature amount may be included in the road surface data. In this situation, the body-side system 2 can also estimate the state of the road surface before the variation by comparing the previous feature amount with the carrier vector. Therefore, it is possible to estimate the state of the road surface before and after the variation and more accurately recognize a variation in the state of the road surface.In the above embodiment, the controller 25 of the receiver 21 in the body-side system 2 evaluates the degree of similarity between the present feature amount and the carrier vector, performs the estimation of the state of the road surface. However, this case is only an example. Other ECUs, e.g., a controller in the brake ECU 22, may determine the degree of similarity and estimate the state of the road surface.

Claims

A tire-side device (1) mounted on a tire (3) included in a vehicle and applied to a tire apparatus (100) for estimating a state of a road surface on which the vehicle travels, the tire-side device (1) comprising: vibration detector (10) configured to output a detection signal according to a vibration level of the tire (3), controller (11) including feature amount extractor (11a) configured to extract a feature amount of the detection signal upon rotation of the tire (3), and transmitter (12) configured to transmit road surface data including the feature amount extracted by the feature amount extractor (11a), the feature amount, which is extracted by the feature amount extraction device (11a) at a past rotation of the tire (3) is a past feature amount, the feature amount extracted by the feature amount extraction device (11a) at a present rotation of the tire (3) is a present feature amount, wherein the control device (11) further includes: a feature amount memory (11b) configured to store the past feature amount, a variation determination device (11c) configured to determine that a variation occurs in the state of the road surface based on the past feature amount and the present feature amount, and a transmission control device (11d) configured to control transmission of the road surface data and cause transmission device (12) to transmit the road surface data including the present feature amount, in response to the variation determining device (11c) determining that the change in the state of the road surface occurs.The tire-side device (1) according to claim 1, wherein the variation determination device (11c) is configured to calculate a degree of similarity between the present feature amount and the past feature amount, and determine that the variation occurs in the state of the road surface based on the degree of similarity.The tire-side device (1) according to claim 1 or 2, wherein: the feature amount extracted by the feature amount extraction device (11a) at a rotation before the present rotation is a previous feature amount, the feature amount storage (11b) is configured to store the previous feature amount, and the variation determination device (11c) is configured to determine that the variation occurs in the state of the road surface based on the present feature amount and the previous feature amount.The tire-side device (1) according to any one of claims 1 to 3, wherein the feature amount extracted by the feature amount extraction device (11a) is expressed by a feature vector having a time axis waveform of the detection signal.A tire apparatus (100) comprising: the tire-side apparatus according to any one of claims 1 to 4, wherein the controller (11) is configured as a first controller; and a body-side system (2) including a receiver (21) configured to receive the road surface data including the present feature amount transmitted from the transmitter (12), and a second controller (25) configured to estimate the state of the road surface based on the road surface data received from the receiver (21), wherein the second controller (25) includes: a carrier vector memory (25a) configured to store the carrier vector of the feature amount for each kind of the state of the road surface, and a state estimation device (25b) configured to:, to estimate the state of the road surface based on the present feature amount included in the road surface data and the carrier vector stored in the carrier vector memory (25a).

Images