Method and device for transmitting data for a vehicle based on anomaly detection during a journey of the vehicle

Abstract

The present invention relates to a method and a device for transmitting data for a first vehicle (11). For this purpose, data representative of journey attributes are obtained by the first vehicle (11) from a set of journey tracking devices on board the first vehicle (11). These data are stored in memory and then supplied to an anomaly detection model to determine information representative of the presence or absence of an anomaly during the journey of the first vehicle (11). The anomaly detection model has been learned from data of attributes of journeys made by a set of second vehicles (12). Depending on the presence or absence of an anomaly, part of the data or all of the data is transmitted to a remote device (110) via a wireless link. Figure for abstract: Figure 1

Description

Title of the invention: Method and device for transmitting data for a vehicle based on an anomaly detection during a journey of the vehicle Technical field

[0001] The invention relates to methods and devices for transmitting data for a vehicle based on a result of anomaly detection during a journey of the vehicle, in particular but not exclusively for a motor vehicle. The invention also relates to a method and a device for detecting anomaly on a journey of a vehicle. The invention also relates to a method and a device for selecting data to be transmitted based on a result of anomaly detection during a journey of the vehicle. Technological background

[0002] The development of connected vehicles, i.e. vehicles configured to communicate data using a wireless communication mode of the OTA type (from the English "Over-The-Air" or in French "by air") with one or more remote locations, for example one or more servers, allows the feedback of information on the journeys made by these vehicles. This information is for example obtained from journey tracking devices, for example sensors, embedded in the vehicles. This information makes it possible, for example, to monitor the operating status of the systems and components embedded in the vehicles and / or to obtain information on the environment in which these vehicles are traveling.

[0003] The volume of data representative of the information acquired by the various journey tracking devices is significant, the transmission of this data using the wireless communication mode leading to significant bandwidth and / or throughput requirements, which can prove costly for the owners or managers of the vehicles.

[0004] To reduce the bandwidth required for the transmission of this data, it is known to temporally sample the acquired data or to temporally space out the acquisition of the data (for example every 1, 10 or 60 seconds) to transmit only part of the data.

[0005] However, such an approach limits the possibilities of analyzing the behavior of vehicles with a risk of not detecting a possible problem or a warning sign of a possible problem at the level of a vehicle. Summary of the present invention

[0006] An object of the present invention is to solve at least one of the problems of the technological background described above.

[0007] Another object of the present invention is for example to improve the collection of vehicle usage data.

[0008] Another object of the present invention is, for example, to optimize the resources necessary for the transmission of vehicle data according to a wireless communication mode.

[0009] Another object of the present invention is, for example, to improve the monitoring of journeys made by a vehicle.

[0010] According to a first aspect, the present invention relates to a method of transmitting data for a first vehicle, the method comprising the following steps: - receiving, from a set of route tracking devices on board the first vehicle, data representative of each attribute of a set of attributes acquired along a current route of the first vehicle; - recording the data in a data storage device of the first vehicle; - determination of information representative of the presence or absence of an anomaly during the current journey by feeding an anomaly detection model learned in a phase known as learning the anomaly detection model from learning data representative of the set of attributes acquired along journeys of a set of second vehicles prior to the current journey; - transmission, for each attribute of the set of attributes, of part of the data or all of the data depending on the information of presence or absence of anomaly to a remote device via a wireless link.

[0011] Determining the presence or absence of detection of an anomaly or a problem during a current journey of the vehicle from data representing a set of journey attributes acquired throughout the journey allows choosing between transmitting all of the acquired data or only a part of the acquired data. This allows transmitting all the data when the situation requires it (for example in the presence of a probable detected anomaly) and thus allows a more complete analysis of the situation from the transmitted data. In other situations (for example in the absence of an anomaly), only a part of the data is transmitted, which reduces the resource requirements for transmitting these data, for example the bandwidth requirements.

[0012] According to a variant, only part of the data is transmitted when the information is representative of an absence of anomaly during the current journey and all of the data is transmitted when the information is representative of a presence of an anomaly during the current journey.

[0013] According to another variant, the method further comprises a step of selecting the part of the data among the data when the information is representative of an absence of anomaly during the current journey, the selection corresponding to a selection of a set of samples representative of the data by temporal sampling of the data.

[0014] According to another variant, the steps of determining the information and transmitting part or all of the data are implemented following the end of the current journey made by the first vehicle.

[0015] According to an additional variant, the anomaly detection model is learned by applying a probability density function of a multivariate Gaussian distribution to the training data.

[0016] According to yet another variant, the presence of an anomaly is detected during the current journey when a probability obtained from the probability density function of a multivariate Gaussian distribution applied to the data is lower than a determined threshold.

[0017] According to another variant, when the information is representative of the presence of an anomaly during the current journey, the method further comprises a step of adjusting a set of parameters of the anomaly detection model as a function of the data.

[0018] According to an additional variant, the set of attributes comprises: - attributes representative of the operation of a set of vehicle components or systems; - representative attributes of the vehicle's exterior environment.

[0019] According to a second aspect, the present invention relates to a data transmission device for a vehicle, the device comprising a memory associated with a processor configured for implementing the steps of the method according to the first aspect of the present invention.

[0020] According to a third aspect, the present invention relates to a vehicle, for example a motor vehicle, comprising a device as described above according to the second aspect of the present invention.

[0021] According to a fourth aspect, the present invention relates to a computer program which comprises instructions adapted for executing the steps of the method according to the first aspect of the present invention, in particular when the computer program is executed by at least one processor.

[0022] Such a computer program may use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0023] According to a fifth aspect, the present invention relates to a support computer-readable recording medium having recorded thereon a computer program comprising instructions for carrying out the steps of the method according to the first aspect of the present invention.

[0024] On the one hand, the recording medium may be any entity or device capable of storing the program. For example, the medium may comprise a storage means, such as a ROM memory, a CD-ROM or a microelectronic circuit type ROM memory, or even a magnetic recording means or a hard disk.

[0025] Furthermore, this recording medium may also be a transmissible medium such as an electrical or optical signal, such a signal being able to be conveyed via an electrical or optical cable, by conventional or hertzian radio or by self-directed laser beam or by other means. The computer program according to the present invention may in particular be downloaded from an Internet-type network.

[0026] Alternatively, the recording medium may be an integrated circuit in which the computer program is incorporated, the integrated circuit being adapted to perform or to be used in performing the method in question. Brief description of the figures

[0027] Other characteristics and advantages of the present invention will emerge from the description of the particular and non-limiting exemplary embodiments of the present invention below, with reference to the appended figures 1 to 3, in which:

[0028] [Fig.l] schematically illustrates a communication environment for connected vehicles, according to a particular exemplary embodiment of the present invention;

[0029] [Fig.2] illustrates a device configured for the transmission of data by a vehicle of [Fig.l], according to a particular and non-limiting exemplary embodiment of the present invention.

[0030] [Fig.3] illustrates a flowchart of the different steps of a method of transmitting data by a vehicle of [Fig.l], according to a particular and non-limiting exemplary embodiment of the present invention. Description of examples of implementation

[0031] A method and a device for transmitting data for a vehicle will now be described in the following with joint reference to FIGS. 1 to 3. The same elements are identified with the same reference signs throughout the description which follows.

[0032] The terms “first(s)”, “second(s)” (or “first(s)”, “second(s)”), etc. are used in this document by arbitrary convention to enable different elements (such as operations, means, etc.) to be identified and distinguished. implemented in the embodiments described below. Such elements may be distinct or correspond to a single element, depending on the embodiment.

[0033] According to a particular and non-limiting example of embodiment of the present invention, the collection and transmission of data for a first vehicle are for example implemented by a device or system embedded in the first vehicle, in particular by one or more processors of one or more computers of the first vehicle. The first vehicle corresponds to a so-called connected vehicle, that is to say a vehicle configured to communicate data to one or more remote devices, for example one or more server-type data processing devices, via a wireless communication network.

[0034] For this purpose, data representative of each attribute of a set of attributes acquired along a current route of the first vehicle by a set of route tracking devices embedded in the first vehicle are received from this or these route tracking devices. A route tracking device corresponds to any device configured for the acquisition of data relating to a route, such as sensors measuring physical quantities representative of the operation of one or more organs or systems of the first vehicle, a communication device or interface receiving data from one or more remote devices, for example environmental data of the first vehicle, or even computers or controllers controlling the on-board systems of the first vehicle. These data are advantageously recorded in a data storage device of the first vehicle, such as a memory or a buffer memory.

[0035] The data stored in memory are provided as input to an anomaly detection or prediction model to determine information representative of the presence or absence of an anomaly during the current journey. Such an anomaly detection or prediction model has been learned during a learning phase prior to a phase of use (or inference) of the model learned from learning data representative of the set of attributes acquired along journeys made by a set of second vehicles prior to the current journey of the first vehicle. The set of second vehicles comprises for example several thousand, tens of thousands or hundreds of thousands of connected second vehicles feeding back the learning data acquired by a set of journey tracking devices embedded in each of these second vehicles.

[0036] Finally, all or part of the data is transmitted to a remote device via a wireless link, according to a wireless communication mode, depending on whether the information is representative of an absence of anomaly or a presence of an anomaly. In other words, part of the data or all of the data associated with each attribute of the set of attributes is transmitted depending on the information of presence or absence of anomaly.

[0037] [Fig.l] schematically illustrates a communication environment 1 of a first vehicle 11 and a set of second vehicles 12 connected to a wireless communication network, according to a particular and non-limiting exemplary embodiment of the present invention.

[0038] The first vehicle 11 corresponds for example to a vehicle with a thermal engine, with electric motor(s) or even a hybrid vehicle with a thermal engine and one or more electric motors. The first vehicle 11 thus corresponds for example to a land vehicle, for example an automobile, a truck, a bus, a motorcycle.

[0039] Each second vehicle 12 is similar to the first vehicle 11 and corresponds to a vehicle with a thermal engine, with an electric motor(s) or even a hybrid vehicle with a thermal engine and one or more electric motors.

[0040] The first vehicle 11 and each second vehicle 12 each correspond to a so-called connected vehicle in that it carries a communication system configured to communicate with one or more remote devices 110, 111 via an infrastructure of a wireless communication network. The remote device 110, 111 advantageously corresponds to a device configured to process data, for example data stored in the memory of the remote device 110, 111 and / or data received from the first and second vehicles 11, 12. Each remote device 110, 111 corresponds for example to a server or a computer of the “cloud” 100.

[0041] The communication system of a connected vehicle comprises, for example, one or more communication antennas connected to a telematic control unit, called TCU (from the English "Telematic Control Unit"), itself connected to one or more computers of the on-board system of the connected vehicle. The antenna(s), the TCU unit and the computer(s) form, for example, a multiplexed architecture for the realization of different services useful for the proper functioning of the connected vehicle and for assisting the driver and / or passengers of the connected vehicle in the control of the connected vehicle and / or for establishing a diagnosis on the functioning of one or more components of the connected vehicle.The computer(s) and the TCU communicate and exchange data with each other via one or more computer buses, for example a communication bus of the data bus type CAN (from the English "Controller Area Network" or in French "Réseau de contrôles"), CAN FD (from the English "Controller Area Network Flexible Data-Rate" or in French "Réseau de contrôles à débit de données flexible"), FlexRay (according to the ISO 17458 standard) or Ethernet (according to the ISO / IEC 802-3 standard).

[0042] The mobile communication infrastructure enabling wireless data communication between a connected vehicle, i.e. the first vehicle 11 and / or each second vehicle 12, and the remote device(s) 110, 111 comprises for example one or more communication devices 101 of the relay antenna type (of a cellular network) or roadside unit, called UBR. In a communication mode using such a network architecture, the data are for example transmitted by the vehicle connected to the remote device 110 of the “cloud” 100 via a relay antenna 101 (the antenna 101 being for example connected to the “cloud” 100 via a wired link and the remote device 110 itself being connected to the network infrastructure of the “cloud” 100 via a wired and / or wireless network).

[0043] The wireless communication system allowing the exchange of data between a connected vehicle corresponding to the first vehicle 11 or to a second vehicle 12 and the remote device(s) 110, 111 corresponds for example to: - a vehicle-to-infrastructure (V2I) communication system, for example based on the 3GPP LTE-V or IEEE 802.1 Ip standards of ITS G5; or - a cellular network type communication system, for example an LTE (Long-Term Evolution) or LTE-Advanced (LTE-Advanced) type network, also called LTE 3G, 4G or 5G.

[0044] A process for transmitting and / or processing data from vehicles connected to the wireless communication network, particular embodiments of which are described below, is advantageously implemented by one or more processors of one or more devices on board the first vehicle 11 or by a system comprising the first vehicle 11, a set of second vehicles 12 and one or more remote devices 110, 111 according to different embodiments.

[0045] A phase prior to the implementation of the data transmission process by the first vehicle 11 corresponds to a so-called learning or training phase of one or more anomaly detection or prediction models from learning data. This learning phase is followed in time by a production or inference phase based on the model(s) learned in the learning phase and data feeding the learned model(s), during which a learned model is used to detect or predict the presence or absence of an anomaly during a journey made by the first vehicle 11 from data collected by the first vehicle 11 during the journey.

[0046] The learning phase is for example implemented by a server-type data processing device, for example the remote device 110 of the “cloud” 100 using learning data collected by the set of second vehicles 12 during journeys made by these second vehicles 12.

[0047] The learning or training of the model implemented in the learning phase corresponds to supervised learning from a set of training data obtained from the set of second vehicles 12, which comprises for example a few tens, a few hundreds, thousands or tens / hundreds of thousands of second vehicles. According to an alternative embodiment, the learning implemented in the first phase corresponds to unsupervised learning from the set of data associated with the set of second vehicles 12.

[0048] According to a particular example, the second vehicles 12 included in the set of second vehicles 12 all have the same configuration or a similar configuration, that is to say that they all correspond to the same type of vehicle (for example the same series of a particular model of vehicle) with identical or similar components or members (same version for example). According to this example, the second vehicles 12 are for example of the same type as the first vehicle 11.

[0049] According to another particular example, the second vehicles 12 included in the set of second vehicles 12 have different configurations. According to such an example, a classification of the second vehicles 12 is for example implemented on the basis of data representative of their type and / or configuration to obtain a set of classes or subgroups each comprising a homogeneous population of second vehicles 12, that is to say second vehicles of the same type and / or same configuration. An anomaly detection model is then for example learned for each subgroup from the learning data obtained from each subgroup.

[0050] According to yet another particular example, the second vehicles 12 included in the set of second vehicles 12 have different configurations. According to this example, a single generic anomaly detection model is learned or trained from the training data obtained from these second vehicles 12, such a generic anomaly detection model being intended to be applied to any type of first vehicle 11.

[0051] In a first operation of the learning phase, the remote device 110 collects learning data representative of a set of attributes or indicators of one or more journeys made by each of the second vehicles 12. The learning data are for example transmitted by each of the second vehicles via the wireless network infrastructure described above to the remote device 110. The learning data correspond for example to data acquired by a set of journey tracking devices on board each of the second vehicles, namely sensors monitoring or measuring parameters representative of the use of components, organs, systems or applications on board each second vehicle 12 and / or communication devices receiving data or parameters from one or more devices remote 111 (for example data relating to the environment in which the second vehicles 12 make the journeys) and / or computers or controllers controlling the on-board systems of each second vehicle. The learning data are for example transmitted at regular intervals, at each acquisition or at the end of each journey (the data then being stored locally and temporarily in a buffer type storage device for transmission at the end of the journey (for example following the stopping of the engine).

[0052] According to an alternative embodiment, one or more processing operations are applied to the data transmitted by the second vehicles 12 to obtain the attributes or indicators of the journeys, the learning data then corresponding to the data obtained after processing(s).

[0053] The attributes or path indicators determined from the collected data include for example: - attributes representative of the operation of a set of vehicle components or systems; and - representative attributes of the vehicle's exterior environment.

[0054] The representative attributes of operation of a set of vehicle components or systems include, for example, for each journey made: - attributes associated with the behavior and / or driving of a vehicle, such as average speed during the trip, average engine speed during the trip, average acceleration during the trip, average number of braking operations per distance traveled, total trip duration, total distance traveled, number of turns, number of lane changes, time spent driving at night, etc.; and / or - attributes associated with the comfort of the vehicle's passengers, such as the interior temperature of the vehicle, the use of elements forming the infotainment system, known as IVI (from the English "In-Vehicle Infotainment" or in French "Infodivertissement monté"), the use of comfort systems such as air conditioning, the heating system and / or seat massage, etc.; and / or - attributes associated with vehicle safety, such as activation of safety systems (e.g. ABS, airbags, etc., automatic emergency braking), frequency of use of turn signals, activation of hazard lights, activation of main or position lights, etc.; and / or - attributes associated with energy consumption (electrical or combustible fuel) and / or energy saving, such as fuel consumption per 100 km, the amount of electrical energy consumed on average, the consumption of electrical energy recovered by regenerative braking, etc.; and / or - attributes associated with vehicle performance, such as maximum speed reached during the journey, maximum engine speed reached during the journey, the maximum longitudinal acceleration reached during the journey, the maximum lateral acceleration reached during the journey, etc.; and / or - attributes associated with the use of ADAS systems (Advanced Driver-Assistance System), such as the usage rate of each ADAS system.

[0055] The representative attributes of the vehicle's exterior environment include, for example, for each journey, attributes associated with the environment and the weather conditions obtained from dedicated servers and / or on-board sensors, such as the exterior temperature, the humidity level in the exterior air, the weather conditions (rain, snow, ice), visibility (fog, precipitation), the presence of traffic jams or slow traffic, the average speed compared to the speed limit applicable on each portion of the road of the journey, the level of pollution of the exterior air (for example the concentration of fine particles), a histogram of time spent per slope class, a histogram of time spent per altitude class, etc.

[0056] In a second operation of the learning phase, the anomaly detection or prediction model(s) is / are generated or trained, for example by classifying the attributes.

[0057] According to a particular exemplary embodiment, the anomaly detection or prediction model is trained using a multivariate Gaussian distribution as anomaly detection model, the model consisting of estimating the parameters of the distribution from the learning data (also called training data), then using this distribution to identify the anomaly(ies) associated with a path.

[0058] The probability density function (PDF) of a multivariate Gaussian distribution is defined as follows:

[0059] [Math.l] 777377777^777'77 <>:SP ( ""MF sr'u - po )

[0060] With: X: is the vector of observed attributes (a vector of characteristics of dimension n), p: is the mean vector of the distribution, E: is the covariance matrix of the distribution, lEl: represents the determinant of the covariance matrix, and n: is the dimension of the data.

[0061] When training the anomaly detection model, the mean p and the covariance matrix E of the training data are calculated from the obtained training data.

[0062] These parameters p and E are representative of a normal behavior or path according to the learning data obtained.

[0063] When receiving data representative of a new journey (or new driving) of a vehicle, the probability according to the multivariate Gaussian distribution for this data is calculated using the PDF formula. If the calculated probability is lower than a determined threshold, this new journey is considered to have at least one anomaly.

[0064] The threshold for the detection of anomalies is for example defined according to a predefined significance level or by means of cross-validation. Thus, if Xnewrouiage corresponds to the data representative of attributes of a new journey, the result of f(Xnewrouiagelp,E) is compared to the determined threshold and if this result is lower than the threshold then the journey is considered to have an anomaly. Otherwise, if the result of the comparison indicates that the result of the calculation of the result of f(XnewUrouiagelp,E) is higher than the determined threshold, then the journey is considered to have no anomaly, the journey being considered to be a normal journey.

[0065] A first class and a second class are for example generated during the learning phase of the anomaly detection model, the output of the detection model corresponding to probabilities that the data provided as input to the model belong to the first class and to the second class respectively, the highest probability giving the class to be associated with the input data. The first class is for example representative of a journey with an anomaly and the second class of a journey without anomaly, that is to say a normal journey.

[0066] The parameters defining the anomaly detection model thus learned or trained are transmitted by the remote device 110 to the first vehicle 11 via the wireless network infrastructure. The first vehicle 11 stores these parameters in a memory associated with one or more computers of the first vehicle 11 for the implementation by this or these computers of the production or inference phase during which a journey made by the first vehicle 11 is analyzed on the basis of data representative of the set of previously defined attributes provided as input to the anomaly detection model learned or trained in the learning phase.

[0067] In a first operation of the production phase, the data representative of each attribute of the set of attributes previously defined with regard to the learning phase are received by the computer of the first vehicle 11 implementing the process. These data are for example received from a set of path tracking devices or from computers controlling these path tracking devices.

[0068] A path tracking device corresponds to one or more of the following devices: a sensor measuring physical quantities representative of the operation of a component or system of the first vehicle 11, a communication device or interface (for example a telematic control unit, called TCU (from the English “Telematic Control Unit”)) receiving data from one or more remote devices, for example environmental data of the first vehicle 11, or even a computer or controller controlling an on-board system or a component or a component of the first vehicle 11.

[0069] These data are received by the computer via one or more data buses of the multiplexed architecture of the first vehicle 11.

[0070] As described above, the path attributes or indicators include, for example: - attributes representative of the operation of a set of vehicle components or systems; and - representative attributes of the vehicle's exterior environment.

[0071] These data are acquired by the set of tracking devices along a current route taken by the first vehicle 11, for example at acquisition frequencies varying from one route tracking device to another or common to several route tracking devices and / or upon detection of a particular event (activation of a component (for example the indicators), reaching of a threshold for a monitored parameter, etc.).

[0072] In a second operation of the production phase, the data received in the first operation of the production phase are recorded in a data storage device of the first vehicle 11, for example a RAM type memory or a buffer memory of the computer implementing the process.

[0073] The data is recorded as it is received by the computer, with an optional time stamp for this data (metadata representative of the time of acquisition or reception are, for example, added to the data in the memory).

[0074] In a third operation of the production phase, information representative of the presence or absence of an anomaly during the current journey of the first vehicle 11 is determined from the data recorded in the memory, by feeding the anomaly detection model learned or trained in the learning phase described previously.

[0075] The determination corresponds for example to a classification of the recorded data, the result of the classification giving the value taken by the information; for example '1' when the result of the classification corresponds to a detection of anomaly (presence of anomaly) and '0' when the result of the classification corresponds to an absence of detection of anomaly (corresponding to a normal journey).

[0076] According to a variant, the determination comprises a calculation of the probability according to the multivariate Gaussian distribution for the recorded data using the PDF formula. If the calculated probability is lower than the determined threshold defined previously, the current journey of the first vehicle 11 is considered to have at least one anomaly (the information takes for example the value '1'). If the calculated probability is higher than the determined threshold defined previously, the current journey of the first vehicle 11 is considered to have no anomaly (the information takes for example the value '0').

[0077] This third operation is for example implemented once the current journey has been completed, for example when the engine of the first vehicle 11 is switched off. A current journey is for example defined as corresponding to a period of driving of the first vehicle 11 between the time corresponding to the starting of the engine (detection of the start command) and the time corresponding to the switching off of the engine (detection of the switching off command).

[0078] The value taken by the information is for example temporarily recorded in the buffer memory and the data representative of the attributes of the current journey stored in the data storage device are then deleted. According to a variant, the data of the current journey are not deleted but these data will be overwritten by the data relating to the following journey made by the first vehicle 11.

[0079] In a fourth operation of the production phase, the first vehicle 11 transmits, for each attribute of the set of attributes, a portion of the data or all of the data depending on the information of presence or absence of an anomaly to the remote device 110 via a wireless link according to a wireless communication mode supported by the infrastructure of the wireless communication network. The quantity or volume of data transmitted thus depends on the information, depending on whether the model has detected or predicted the occurrence of one or more anomalies during the current journey or not.

[0080] The transmission is for example controlled by the computer implementing the process via the TCU unit of the first vehicle 11.

[0081] Thus, when the information is representative of the absence of anomaly during the current journey, only part of the data stored in the memory of the first vehicle 11 is transmitted. Indeed, in the absence of an anomaly, the journey is judged to be normal and it does not prove necessary to transmit all the data, a detailed analysis of all the data not appearing necessary. This makes it possible to save the resources necessary for the communication of such data, such as the bandwidth allocated by the network but also the computing power necessary on the side of the first vehicle 11 to control the transmission and the memory footprint necessary on the side of the remote device 110 to store the data thus received.

[0082] When the information is representative of the presence of an anomaly during the current journey, all of the data stored in the memory of the first vehicle 11 is transmitted. Indeed, in the absence of an anomaly, the journey is judged to be abnormal and the remote device 110 may need all of the data for analysis and detection of the problem(s) encountered during the current journey.

[0083] When the information is representative of the absence of anomaly during the current journey, the part of the data transmitted is obtained by selection from the set of data stored in memory of the part of the data to be transmitted, for each attribute of the set of attributes.

[0084] The selection is for example made randomly to obtain a set of samples representative of all the data representative of each attribute, with for example a maximum limit of volume or quantity of data to be transmitted for each attribute.

[0085] According to another example, the selection is obtained by applying a temporal sampling to the data representative of each attribute, with a determined sampling rate. The sampling rate is for example identical for all the attributes or varies according to the nature or type of the attributes.

[0086] Such temporal sampling makes it possible to select a set of samples representative of all the data representative of each attribute, by selecting the data at determined temporal intervals, for example the data acquired every 250, 500, 1000, 5000 or 10000 ms.

[0087] According to a particular embodiment, when the information is representative of the presence of an anomaly during the current journey, the process further comprises a new training of the anomaly detection model on the basis of the data associated with the current journey of the first vehicle 11 received in their entirety. The new training corresponds to an adjustment of the parameters of the anomaly detection model according to the received data. The new training is for example implemented by the remote device, and this for each time that a connected vehicle transmits to it data associated with the detection of an anomaly during a journey.

[0088] Once the parameters have been recalculated, the latter are for example transmitted to the first vehicle 11 by the remote device 110 via the infrastructure of the wireless communication network to update the detection model stored in the memory of the first vehicle 11. This makes it possible for example to adjust the detection of an anomaly based on the ground feedback from the vehicles in circulation, an abnormal situation for the second vehicles 12 being able to become normal over time (for example in the event of a change in the circulation environment of the vehicles).

[0089] [Fig.2] schematically illustrates a device 2 configured for communication and / or processing of vehicle data, according to a particular exemplary embodiment. and not limiting of the present invention.

[0090] The device 2 corresponds for example to a device on board a vehicle, for example in the first vehicle 11, such as a computer or a TCU unit.

[0091] According to another exemplary embodiment, the device 2 corresponds to a calculation device or a data processing device such as the remote device 110, for example a computer or a server.

[0092] The device 2 is for example configured for the implementation of the operations described with regard to [Fig.l] and / or the steps of the method described with regard to [Fig.3]. Examples of such a device 2 include, but are not limited to, a computer, a laptop, a server, on-board electronic equipment such as an on-board computer of a vehicle, an electronic calculator such as an ECU (“Electronic Control Unit”), a smartphone, a tablet. The elements of the device 2, individually or in combination, can be integrated in a single integrated circuit, in several integrated circuits, and / or in discrete components. The device 2 can be produced in the form of electronic circuits or software (or computer) modules or even a combination of electronic circuits and software modules.

[0093] The device 2 comprises one (or more) processor(s) 20 configured to execute instructions for carrying out the steps of the method and / or for executing the instructions of the software(s) embedded in the device 2. The processor 20 may include integrated memory, an input / output interface, and various circuits known to those skilled in the art. The device 2 further comprises at least one memory 21 corresponding for example to a volatile and / or non-volatile memory and / or comprises a memory storage device which may comprise volatile and / or non-volatile memory, such as EEPROM, ROM, PROM, RAM, DRAM, SRAM, flash, magnetic or optical disk.

[0094] The computer code of the embedded software(s) comprising the instructions to be loaded and executed by the processor is for example stored in the memory 21.

[0095] According to various particular and non-limiting embodiments, the device 2 is coupled in communication with other similar devices or systems and / or with communication devices, for example a TCU (from the English “Telematic Control Unit” or in French “Telematic Control Unit”), for example via a communication bus or through dedicated input / output ports.

[0096] According to a particular and non-limiting exemplary embodiment, the device 2 comprises a block 22 of interface elements for communicating with external devices. The interface elements of the block 22 comprise one or more of the following interfaces: - RF radio frequency interface, for example Wi-Fi® type (according to IEEE 802.11), for example in the 2.4 or 5 GHz frequency bands, or Bluetooth® type (according to IEEE 802.15.1), in the 2.4 GHz frequency band, or Sigfox type using UBN (Ultra Narrow Band) radio technology, or LoRa in the 868 MHz frequency band, LTE (LTE) “Long-Term Evolution” or in French “Long-Term Evolution”), LTE-Advanced (or in French LTE-advanced); - USB interface (from the English “Universal Serial Bus” or “Universal Serial Bus” in French); - HDMI interface (from the English “High Definition Multimedia Interface” or “High Definition Multimedia Interface” in French); - LIN interface (from the English “Local Interconnect Network”).

[0097] According to another particular and non-limiting exemplary embodiment, the device 2 comprises a communication interface 23 which makes it possible to establish communication with other devices (such as other computers of the on-board system) via a communication channel 230. The communication interface 23 corresponds for example to a transmitter configured to transmit and receive information and / or data via the communication channel 230. The communication interface 23 corresponds for example to a wired network of the Ethernet type (standardized by the ISO / IEC 802-3 standard).

[0098] According to a particular and non-limiting exemplary embodiment, the device 2 can provide output signals to one or more external devices, such as a display screen 240, touch-sensitive or not, one or more speakers 250 and / or other peripherals 260 (projection system) via output interfaces 24, 25 and 26 respectively. According to a variant, one or other of the external devices is integrated into the device 2.

[0099] [Fig. 3] illustrates a flowchart of the different steps of a data transmission method for a first vehicle, for example the first vehicle 11, according to a particular and non-limiting exemplary embodiment of the present invention. The method is for example implemented by one or more processors of one or more computers of the first vehicle, or by the device 2 of [Fig. 2].

[0100] In a first step 31, data representative of each attribute of a set of attributes acquired along a current path of the first vehicle are received from a set of path tracking devices on board the first vehicle.

[0101] In a second step 32, the data received in the first step 31 are recorded in a data storage device of the first vehicle.

[0102] In a third step 33, information representative of the presence or absence of an anomaly during the current journey is determined by feeding an anomaly detection model learned in a phase called learning the anomaly detection model from learning data representative of the set of attributes acquired along journeys of a set of second vehicles prior to the current journey.

[0103] In a fourth step 34, part of the data or all of the data is transmitted, for each attribute of the set of attributes, depending on the information of presence or absence of anomaly to a remote device via a wireless link.

[0104] According to a variant, the variants and examples of the operations described in relation to [Fig.l] apply to the steps of the method of [Fig.3].

[0105] Of course, the present invention is not limited to the exemplary embodiments described above but extends to a method for learning a model for detecting or predicting an anomaly associated with a vehicle journey without thereby departing from the scope of the present invention. The same would apply to a device configured for implementing such a method.

Claims

Claims

1. A method of transmitting data for a first vehicle (11), said method comprising the following steps: - receiving (31), from a set of route tracking devices on board said first vehicle (11), data representative of each attribute of a set of attributes acquired along a current route of said first vehicle (11); - recording (32) said data in a data storage device of said first vehicle (11); - determining (33) information representative of the presence or absence of an anomaly during said current route by feeding an anomaly detection model learned in a so-called learning phase of said anomaly detection model from learning data representative of said set of attributes acquired along routes of a set of second vehicles (12) prior to said current route;- transmission (34), for each attribute of said set of attributes, of a part of said data or of all of said data as a function of said information of presence or absence of anomaly to a remote device via a wireless link.;

2. Method according to claim 1, for which only a part of said data is transmitted when said information is representative of an absence of anomaly during said current journey and all of said data is transmitted when said information is representative of a presence of anomaly during said current journey.

3. A method according to claim 2, further comprising a step of selecting said portion of said data from among said data when said information is representative of an absence of anomaly during said current path, said selection corresponding to a selection of a set of samples representative of said data by time sampling of said data.

4. Method according to one of claims 1 to 3, for which said steps of determining said information and transmitting part or all of said data are implemented following an end of said current journey made by said first vehicle (11).

5. Method according to one of claims 1 to 4, for which said anomaly detection model is learned by applying a function of probability density of a multivariate Gaussian distribution to said training data.

6. Method according to claim 5, for which the presence of an anomaly is detected during said current path when a probability obtained from said probability density function of a multivariate Gaussian distribution applied to said data is lower than a determined threshold.

7. Method according to one of claims 1 to 6, for which, when said information is representative of a presence of an anomaly during said current journey, the method further comprises a step of adjusting a set of parameters of said anomaly detection model as a function of the data.

8. Method according to one of claims 1 to 7, for which said set of attributes comprises: - attributes representative of the operation of a set of vehicle components or systems; - attributes representative of the vehicle's external environment.

9. Device (3) for transmitting data for a vehicle, said device (3) comprising a memory (31) associated with at least one processor (30) configured for implementing the steps of the method according to any one of claims 1 to 8.

10. Vehicle comprising the device according to claim 9.

Images